ArticlePDF Available

Medics as a Positive Deviant in Influenza Vaccination: The Role of Vaccine Beliefs, Self-Efficacy and Contextual Variables

Authors:

Abstract and Figures

The influenza vaccination rate remains unsatisfactorily low, especially in the healthy adult population. The positive deviant approach was used to identify key psychosocial factors explaining the intention of influenza vaccination in medics and compare them with those in non-medics. Methods: There were 709 participants, as follows: 301 medics and 408 non-medics. We conducted a cross-sectional study in which a multi-module self-administered questionnaire examining vaccination beliefs, risk perception, outcome expectations (gains or losses), facilitators' relevance, vaccination self-efficacy and vaccination intention was adopted. We also gathered information on access to vaccination, the strength of the vaccination habit and sociodemographic variables. Results: We used SEM and were able to explain 78% of the variance in intention in medics and 56% in non-medics. We identified both direct and indirect effects between the studied variables. In both groups, the intention was related to vaccination self-efficacy, stronger habits and previous season vaccination, but access to vaccines was significant only in non-medics. Conclusions: Applying the positive deviance approach and considering medics as positive deviants in vaccination performance extended the perspective on what factors to focus on in the non-medical population. Vaccination promotion shortly before the flu season should target non- or low-intenders and also intenders by the delivery of balanced information affecting key vaccination cognitions. General pro-vaccine beliefs, which may act as implicit attitudes, should be created in advance to build proper grounds for specific outcome expectations and facilitators' recognition. It should not be limited only to risk perception. Some level of evidence-based critical beliefs about vaccination can be beneficial.
Content may be subject to copyright.
Citation: Włodarczyk, D.;
Zi˛etalewicz, U. Medics as a Positive
Deviant in Influenza Vaccination: The
Role of Vaccine Beliefs, Self-Efficacy
and Contextual Variables. Vaccines
2022,10, 723. https://doi.org/
10.3390/vaccines10050723
Academic Editors: Tokiko
Watanabe, Shinji Watanabe
and Giuseppe La Torre
Received: 10 March 2022
Accepted: 29 April 2022
Published: 5 May 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
Medics as a Positive Deviant in Influenza Vaccination: The Role
of Vaccine Beliefs, Self-Efficacy and Contextual Variables
Dorota Włodarczyk and Urszula Zi˛etalewicz *
Department of Health Psychology, Medical University of Warsaw, 00-575 Warsaw, Poland;
dorota.wlodarczyk@wum.edu.pl
*Correspondence: urszula.zietalewicz@wum.edu.pl
Abstract:
The influenza vaccination rate remains unsatisfactorily low, especially in the healthy adult
population. The positive deviant approach was used to identify key psychosocial factors explain-
ing the intention of influenza vaccination in medics and compare them with those in non-medics.
Methods: There were 709 participants, as follows: 301 medics and 408 non-medics. We conducted a
cross-sectional study in which a multi-module self-administered questionnaire examining vaccination
beliefs, risk perception, outcome expectations (gains or losses), facilitators’ relevance, vaccination
self-efficacy and vaccination intention was adopted. We also gathered information on access to
vaccination, the strength of the vaccination habit and sociodemographic variables. Results: We used
SEM and were able to explain 78% of the variance in intention in medics and 56% in non-medics.
We identified both direct and indirect effects between the studied variables. In both groups, the
intention was related to vaccination self-efficacy, stronger habits and previous season vaccination, but
access to vaccines was significant only in non-medics. Conclusions: Applying the positive deviance
approach and considering medics as positive deviants in vaccination performance extended the
perspective on what factors to focus on in the non-medical population. Vaccination promotion shortly
before the flu season should target non- or low-intenders and also intenders by the delivery of
balanced information affecting key vaccination cognitions. General pro-vaccine beliefs, which may
act as implicit attitudes, should be created in advance to build proper grounds for specific outcome
expectations and facilitators’ recognition. It should not be limited only to risk perception. Some level
of evidence-based critical beliefs about vaccination can be beneficial.
Keywords: psychosocial factors; influenza vaccination; medics; non-medics
1. Introduction
According to the World Health Organization [
1
], the influenza vaccine is an effective
prevention against seasonal influenza and is recommended for healthy adults and specific
risk groups. Despite the availability of safe vaccines and influenza’s extensive outcomes,
vaccine uptake in the general public and high-risk groups is rather low. In Europe [2], the
median for healthcare workers was 30.2%. For example, in Poland, 4.1% of the population
was vaccinated against influenza in the season 2019/2020 [
3
], which was one of the lowest
flu vaccination rates, similar to previously reported seasons [2].
To determine factors that increase influenza vaccine uptake in the general public, we
used positive deviance, an asset-based approach that learns from those who demonstrate
better performance in an outcome of interest by identifying their strengths and resources
and exploring how and why things go right [
4
]. Although the implementation of influenza
vaccinations among medics is not satisfactory, it is higher than in the general public (non-
medics), and this group can potentially be treated as a positive deviant.
The two groups obviously differ in the type of education, but both consist of healthy
adults who not only are at risk of infection and its outcomes but also play an important
role in virus transmission [
1
]. It was confirmed that medics are not free from vaccination
Vaccines 2022,10, 723. https://doi.org/10.3390/vaccines10050723 https://www.mdpi.com/journal/vaccines
Vaccines 2022,10, 723 2 of 13
hesitancy [
5
,
6
], and, paradoxically, being the most trusted information sources by the public,
some of them are losing confidence in vaccines [7].
A recent systematic review showed differences in vaccination barriers between the risk
groups and little contribution from most sociodemographic variables [
8
]. This confirmed
the predictive power of selected psychosocial variables, which allowed effective, evidence-
informed interventions to be designed. Studies also identified factors specific to medics [
9
].
The most influential theories used in studies determining vaccination intention and
performance are the health belief model [
10
], the theory of planned behaviour [
11
], and
social cognitive theories [
12
] or the Health Action Process Approach (HAPA; [
13
]). Al-
though there is considerable overlap between them and they differ slightly in terminology,
they indicate key relevant constructs and assume direct and indirect relationships between
them [
14
]. In stage models, it is expected that different factors play a role, depending on
the stage of change [15].
In line with these theories, many studies confirmed that risk perception predicted
vaccination uptake [
16
]. Usually, it addresses the relative vulnerability and compares the
risk of having the flu or negative consequences of the disease compared to people of the
same age and gender [
2
,
17
]. Perceiving a low risk of the disease was identified as a barrier
to influenza vaccine uptake [
8
], suggesting that this factor is rather stage-specific and plays
a role when the motivation to change develops but is insufficient to enable a person to
change behaviour [18].
A number of studies referred to attitudes and beliefs towards vaccinations, showing
that misconceptions, hesitancy or an antivaccination approach were associated with poor
performance [
19
]. This can be considered at the following two levels: general beliefs
(positive or negative) include convictions about influenza, understanding of the safety
and effectiveness of vaccines, unique theories about the purposes of vaccination, civil
obligations and liberties or trust in scientific authorities [
20
]; specific outcome expectancies
include individuals’ perception of links between action and subsequent outcomes and the
specific gains and losses resulting from vaccination [18].
Another factor proved to significantly impact influenza vaccination intention [
21
]
and implementation [
22
] was self-efficacy, which refers to individuals’ beliefs in their
capabilities to exercise control over new behaviour and their own functioning even in the
face of barriers [18].
It was shown that social and environmental facilitators, such as encouragement by
supervisors or well-organized vaccination campaigns with on-site vaccination [
23
] and
receiving reminders to vaccinate (e.g., a text message), had statistically relevant effects on
increasing influenza vaccination rates [24].
Some contextual factors also played a role. The previous seasonal influenza vaccination
was found to be a positive predictor of vaccine uptake [
25
,
26
]. Many research results, also
in medics, confirmed the positive effects of access to vaccination but this can be meaningless
when individuals have to pay for it, especially in poorer countries [27].
Drawing from the theories of health behaviour and the existing results on influenza
vaccination determinants, we selected the key factors related to the vaccination intention in
order to compare them between medics and non-medics. Following HAPA [
28
], we focused
on the paths connecting self-efficacy with a behavioural intention on the one hand and risk
perception and various beliefs about vaccines on the other hand. The conceptual model
tested in the study is presented in Figure 1. We assumed that the different interrelated types
of vaccine beliefs may influence self-efficacy and can give way to the intention [
18
,
29
]. In
line with the attitude concept [
30
,
31
], we focused on more general and more specific beliefs
about influenza vaccination, which constituted ‘distal’ and ‘proximal’ antecedents of self-
efficacy. The general pro- and antivaccine beliefs refer to convictions about flu vaccination
in general without personal commitment. The influenza vaccination-specific beliefs assume
personal commitment and refer to perceived gains and/or losses resulting from vaccination
(also labelled as outcome expectations) and perceived facilitators and/or barriers.
Vaccines 2022,10, 723 3 of 13
Vaccines 2022, 10, x FOR PEER REVIEW 3 of 14
losses resulting from vaccination (also labelled as outcome expectations) and perceived
facilitators and/or barriers.
Figure 1. The conceptual model tested in the study.
The aim of the study was to test the model in medics and non-medics and compare
the effects between the groups. We hypothesized the following: (1) general pro- and anti-
vaccine beliefs together with risk perception would be related to self-efficacy through in-
fluenza vaccination-specific beliefs; (2) self-efficacy would mediate the relationships be-
tween vaccine beliefs and behavioural intention; (3) behavioural intention would be re-
lated also to previous flu vaccination experiences and access to vaccines; (4) the relation-
ships between variables in the model will differ between medics and non-medics.
2. Materials and Methods
2.1. Participants
The results were obtained from 709 participants forming two groups: medics (n =
301) and non-medics (n = 408). They were recruited from the voluntary registered panel
participants who gave informed consent before participating in a study and met the crite-
ria of inclusion. Medics were significantly older than non-medics (M = 52.10, SD = 12.43
vs. M = 41.04, SD = 13.64; t(676, 24) = 11.23, p < 0.001). The number of female participants was
higher among medics (66.1% vs. 50.7%; χ2(1, n = 709) = 16.74, p < 0.001). There was also a sig-
nificant difference in the distribution of education (χ2(2, n = 709) = 212.00, p < 0.001). All medics
had higher education at the level of at least Master’s degree (82.7.1%) or Bachelor’s degree
(17.3%). Among non-medics, the majority had secondary education (40.4%), followed by
primary or vocational education (29.7%) and higher education (29.9%). The groups dif-
fered also in size of living location (χ2(4, n = 709) = 21.62, p < 0.001). Among non-medics, in-
habitants of rural areas (34.1%) and medium-sized towns (28.7%) were more numerous
than inhabitants of other areas. Among medics, the distribution was quite even. Medics
performed the following professions: doctor (n = 151, 50.2%), nurse (n = 106, 35.2%), phys-
iotherapist (n = 4, 1.3%), midwife (n = 7, 2.3%) and paramedic (n = 33, 11.0%).
2.2. Measures
The study instrument was a multi-module self-administered questionnaire con-
structed according to recommendations in the field [32,33]. The questions were built ac-
cording to an algorithm; these are strictly structured sentences in which the part concern-
ing a given type of behaviour can be changed (the translated version of the questionnaire
is available in the Supplementary File S1). Content validity was established by sending
the questionnaire to a panel of five experts. Questions with a mean standard deviation of
experts’ opinions greater than 1.1 were removed. Reliability was confirmed using
Cronbach’s alpha and confirmatory factor analysis (CFA) for each module [34].
Figure 1. The conceptual model tested in the study.
The aim of the study was to test the model in medics and non-medics and compare
the effects between the groups. We hypothesized the following: (1) general pro- and an-
tivaccine beliefs together with risk perception would be related to self-efficacy through
influenza vaccination-specific beliefs; (2) self-efficacy would mediate the relationships be-
tween vaccine beliefs and behavioural intention; (3) behavioural intention would be related
also to previous flu vaccination experiences and access to vaccines; (4) the relationships
between variables in the model will differ between medics and non-medics.
2. Materials and Methods
2.1. Participants
The results were obtained from 709 participants forming two groups: medics
(n= 301)
and non-medics (n= 408). They were recruited from the voluntary registered panel par-
ticipants who gave informed consent before participating in a study and met the criteria
of inclusion. Medics were significantly older than non-medics (M = 52.10, SD = 12.43 vs.
M = 41.04
, SD = 13.64; t
(676, 24)
= 11.23, p< 0.001). The number of female participants was
higher among medics (66.1% vs. 50.7%;
χ2(1, n= 709)
= 16.74, p< 0.001). There was also a
significant difference in the distribution of education (
χ2(2, n= 709)
= 212.00, p< 0.001). All
medics had higher education at the level of at least Master’s degree (82.7.1%) or Bache-
lor’s degree (17.3%). Among non-medics, the majority had secondary education (40.4%),
followed by primary or vocational education (29.7%) and higher education (29.9%). The
groups differed also in size of living location (
χ2(4, n= 709)
= 21.62, p< 0.001). Among non-
medics, inhabitants of rural areas (34.1%) and medium-sized towns (28.7%) were more
numerous than inhabitants of other areas. Among medics, the distribution was quite even.
Medics performed the following professions: doctor (n= 151, 50.2%), nurse (n= 106, 35.2%),
physiotherapist (n= 4, 1.3%), midwife (n= 7, 2.3%) and paramedic (n= 33, 11.0%).
2.2. Measures
The study instrument was a multi-module self-administered questionnaire constructed
according to recommendations in the field [
32
,
33
]. The questions were built according to an
algorithm; these are strictly structured sentences in which the part concerning a given type
of behaviour can be changed (the translated version of the questionnaire is available in the
Supplementary File S1). Content validity was established by sending the questionnaire
to a panel of five experts. Questions with a mean standard deviation of experts’ opinions
greater than 1.1 were removed. Reliability was confirmed using Cronbach’s alpha and
confirmatory factor analysis (CFA) for each module [34].
The module measuring general vaccination beliefs has two subscales: pro-vaccine
beliefs (VacPros) comprising three items on vaccine security, trust in science and pro-social
attitude (sample item: It is safer to be vaccinated against the flu than to get it); antivaccine be-
liefs (VacCons) comprising three items on undesirable effects, low seriousness/prevalence
of influenza and dishonesty of the vaccine industry (sample item: The harmfulness of the
side effects of influenza vaccinations is greater than the resulting benefits). The responses
Vaccines 2022,10, 723 4 of 13
were rated on a five-point scale (1 = strongly disagree; 5 = strongly agree). Reliability of
the VacPros subscale is
α
= 0.80 for medics and
α
= 0.89 for non-medics; for the VacCons
subscale, reliability is
α
= 0.76 for medics and
α
= 0.81 for non-medics. Indices of model fit
in CFA were the following: RMSEA = 0.044, SRMS = 0.019 and CFI = 0.994.
The module measuring risk perception (Risk) refers to relative risk of having flu or
its consequences. It contains four items in which participants rate their individual risk
compared to others on a five-point scale from 1 (definitely lower risk than others) to 5
(definitely higher risk than others). Sample item: Compared to others of the same sex and
age as you, how would you estimate the likelihood that in your current situation you will
have the flu. Reliability for medics is
α
= 0.89 and for non-medics is
α
= 0.89. CFA indices
are the following: RMSEA = 0.076, SRMR = 0.007 and CFI = 0.983.
The module measuring outcome expectations encompasses perceived gains (VacGains)
and perceived losses (VacLosses). The VacGains subscale consists of eight items on per-
ceived gains from influenza vaccination with a five-point rating scale from 1 (strongly
disagree) to 5 (strongly agree). Sample item: Influenza vaccination reduces my risk of
getting the virus and/or getting the flu. Reliability is
α
= 0.93 for medics and
α
= 0.94 for
non-medics. CFA indices are the following: RMSEA = 0.079, SRMR = 0.023 and CFI = 0.982.
The VacLosses subscale consists of nine items measuring perceived losses from influenza
vaccination with a five-point rating scale (1 = strongly disagree; 5 = strongly agree). Sample
item: My relatives would criticize me or be dissatisfied with me. Reliability for medics is
α
= 0.80 and for non-medics is
α
= 0.88. CFA indices are the following: RMSEA = 0.072,
SRMR = 0.038 and CFI = 0.961.
The module measuring facilitators’ relevance (FacRel) includes factors potentially
conducive to influenza vaccination. The scale consists of nine items on the following
question: Regardless of whether you vaccinate or not, does or could this factor favour your
decision to vaccinate? Sample item: Publicly available information to remind you when
and how to vaccinate. The responses are rated on a five-point scale (1 = definitely irrelevant
to me; 5 = definitely relevant to me). Reliability for medics is
α
= 0.89 and for non-medics
is
α
= 0.94. CFA indices are the following: RMSEA = 0.071, SRMR = 0.026 and CFI = 0.977.
The vaccination self-efficacy scale (VacSE) consists of six items on a person’s belief
in the extent to which they are able to perform vaccination, even though difficulties arise.
A five-point scale was used (1 = definitely not sure; 5 = definitely sure). Sample item: To
what extent are you sure that you will vaccinate in the current season even if you have to
pay in full or in part for the influenza vaccination? Reliability for medics is
α
= 0.98 and for
non-medics is
α
= 0.97. CFA indices are the following: RMSEA = 0.078, SRMR = 0.007 and
CFI = 0.994.
The vaccination intention (Intention) was established by asking if a person had an
intention to vaccinate in the current season, with answers selected from the following: I
have already vaccinated (vaccination implementation) to I am not going to get vaccinated.
Additionally, data on the contextual variables (last year’s influenza vaccination, five-year
vaccination, vaccine access) and age, gender and education were collected.
2.3. Procedure
A cross-sectional study was conducted between October and December 2020 (during
the second wave of the COVID-19 pandemic in Poland) by a professional survey company.
The company secured selection of the participants from those who freely registered with
the panel and gave necessary (true and up-to-date) information to participate in studies
run by the company together with their informed consent. Non-medics filled in the above
multi-module survey online. Medics participated in the computer-assisted telephone
interview (CATI), chosen as the method for securing the highest participation rate at
that time (lockdown with low number of patients in facilities and not frequently used
teleconsultations joint with low computer proficiency in medics). The inclusion criteria for
medics were the following: being a professionally active family doctor or pediatrician (50%
Vaccines 2022,10, 723 5 of 13
of the group); or a nurse, midwife, paramedic or physiotherapist; for non-medics, meeting
the age criteria representative of the general public in the country.
2.4. Statistical Analysis
The statistical analysis was conducted using IBM SPSS Statistics 26 and Mplus for
the SEM. The a priori sample size for structural equation modelling (SEM) was calculated
based on the following assumptions [
35
]: anticipated effect size = 0.3, statistical power
level = 0.8, number of latent variables = 45, number of observed variables = 13, probability
level = 0.05 [
36
,
37
]. It was also based on the sample-to-variable ratio, which suggests an
observation-to-variable ratio of 15:1 or 20:1 [
38
]. Regardless of the method of estimation,
the achieved sample size in each group was sufficient.
The first step of the analysis was establishing the equivalence of scales between medics
and non-medics by measuring both metric and scalar invariance [
39
]. The procedure
is highly recommended when different social groups or groups tested under different
conditions are compared (e.g., interview versus online). The criteria of invariance were as
follows [
40
]: change in CFI
≤−
0.010 paired with RMSEA
0.015 or SRMR
0.030 (metric
invariance); change in CFI
≤−
0.010 paired with RMSEA
0.015 or SRMR
0.010 (scalar
invariance). All scales proved to be equivalent and proper for further comparisons between
medics and non-medics.
Descriptive statistics were used to characterize the sample. Chi-square tests were per-
formed to examine group differences in frequencies. Student’s t-test or the Mann–Whitney
U test was used to compare normally or non-normally distributed variables, respectively.
To examine correlations between variables we used Pearson’s coefficient. In order to verify
differences in relationships between predictors of vaccination intention in medics and
non-medics, we used multi-group SEM [
41
,
42
]. The results for both groups are presented
in Figure 2, with ovals and rectangles representing latent and observable variables, re-
spectively, and values of path coefficients located above (for non-medics) and below (for
medics) the arrows representing the strength of the relationships between variables.
Vaccines 2022, 10, x FOR PEER REVIEW 5 of 14
with the panel and gave necessary (true and up-to-date) information to participate in stud-
ies run by the company together with their informed consent. Non-medics filled in the
above multi-module survey online. Medics participated in the computer-assisted tele-
phone interview (CATI), chosen as the method for securing the highest participation rate
at that time (lockdown with low number of patients in facilities and not frequently used
teleconsultations joint with low computer proficiency in medics). The inclusion criteria
for medics were the following: being a professionally active family doctor or pediatrician
(50% of the group); or a nurse, midwife, paramedic or physiotherapist; for non-medics,
meeting the age criteria representative of the general public in the country.
2.4. Statistical Analysis
The statistical analysis was conducted using IBM SPSS Statistics 26 and Mplus for the
SEM. The a priori sample size for structural equation modelling (SEM) was calculated
based on the following assumptions [35]: anticipated effect size = 0.3, statistical power
level = 0.8, number of latent variables = 45, number of observed variables = 13, probability
level = 0.05 [36,37]. It was also based on the sample-to-variable ratio, which suggests an
observation-to-variable ratio of 15:1 or 20:1 [38]. Regardless of the method of estimation,
the achieved sample size in each group was sufficient.
The first step of the analysis was establishing the equivalence of scales between med-
ics and non-medics by measuring both metric and scalar invariance [39]. The procedure
is highly recommended when different social groups or groups tested under different
conditions are compared (e.g., interview versus online). The criteria of invariance were as
follows [40]: change in CFI ≤−0.010 paired with RMSEA 0.015 or SRMR 0.030 (metric
invariance); change in CFI ≤−0.010 paired with RMSEA 0.015 or SRMR 0.010 (scalar in-
variance). All scales proved to be equivalent and proper for further comparisons between
medics and non-medics.
Descriptive statistics were used to characterize the sample. Chi-square tests were per-
formed to examine group differences in frequencies. Student’s t-test or the Mann–Whit-
ney U test was used to compare normally or non-normally distributed variables, respec-
tively. To examine correlations between variables we used Pearson’s coefficient. In order
to verify differences in relationships between predictors of vaccination intention in medics
and non-medics, we used multi-group SEM [41,42]. The results for both groups are pre-
sented in Figure 2, with ovals and rectangles representing latent and observable variables,
respectively, and values of path coefficients located above (for non-medics) and below
(for medics) the arrows representing the strength of the relationships between variables.
Figure 2.
The result of SEM in medics (n= 301; path coefficients in italics below the arrows) and
non-medics (n= 408; path coefficients in regular font above the arrows) analyzed for each group
separately (showed in one figure for better visualization). Note: *** = p< 0.001; ** = p< 0.01;
* = p< 0.05; # = p= 0.05; ns = not significant.
3. Results
3.1. Differences between Groups in Studied Variables
The frequencies of responses reflecting readiness for influenza vaccination are pre-
sented in Figure 3. Medics had higher readiness for getting vaccinated than non-medics
(Me = 1.00 and Me = 5.00, respectively, U = 29928.00 and p< 0.001 after excluding subjects
with contraindications).
Vaccines 2022,10, 723 6 of 13
Vaccines 2022, 10, x FOR PEER REVIEW 6 of 14
Figure 2 The result of SEM in medics (n = 301; path coefficients in italics below the arrows) and non-
medics (n = 408; path coefficients in regular font above the arrows) analyzed for each group sepa-
rately (showed in one figure for better visualization).
Note: *** = p < 0.001; ** = p < 0.01; * = p < 0.05.
3. Results
3.1. Differences between Groups in Studied Variables
The frequencies of responses reflecting readiness for influenza vaccination are pre-
sented in Figure 3. Medics had higher readiness for getting vaccinated than non-medics
(Me = 1.00 and Me = 5.00, respectively, U = 29928.00 and p < 0.001 after excluding subjects
with contraindications).
At the end of December 2020, 48% of medics were vaccinated (versus 8% of non-
medics) and the same percentage of non-medics declared they were not going to get vac-
cinated (versus 24% of medics).
Figure 3. Readiness for influenza vaccination in medics (n = 301) and non-medics (n = 408).
The groups differed significantly in the levels of predictors included in the study,
with the exception of Risk (Table 1). Medics were higher in VacPros, VacGains, FacRel
and VacSE and lower in VacCons and VacLosses. Risk below three indicates a slight bias
towards defensive optimism in both groups.
Table 1. Comparison of predictors in medics (n = 301) and non-medics (n = 408).
Variable Group M (SD) t df p
Pro-vaccine beliefs Medics 4.44 (0.85) 15.10 702.86 <0.001
Non-medics 3.36 (1.06)
Antivaccine beliefs Medics 1.67 (0.89) 15.33 681.59 <0.001
Non-medics 2.76 (1.00)
Risk perception Medics 2.62 (0.88) 0.94 707 0.346
Non-medics 2.68 (0.83)
Perceived gains Medics 4.16 (0.89) 12.55 706 <0.001
Non-medics 3.25 (0.93)
Perceived losses Medics 2.10 (0.82) 14.24 642.14 <0.001
Non-medics 2.99 90.81)
Facilitators’ relevance Medics 3.36 (1.06) 4.30 632.35 <0.001
Figure 3. Readiness for influenza vaccination in medics (n= 301) and non-medics (n= 408).
At the end of December 2020, 48% of medics were vaccinated (versus 8% of non-
medics) and the same percentage of non-medics declared they were not going to get
vaccinated (versus 24% of medics).
The groups differed significantly in the levels of predictors included in the study, with
the exception of Risk (Table 1). Medics were higher in VacPros, VacGains, FacRel and VacSE
and lower in VacCons and VacLosses. Risk below three indicates a slight bias towards
defensive optimism in both groups.
Table 1. Comparison of predictors in medics (n= 301) and non-medics (n= 408).
Variable Group M (SD) t df p
Pro-vaccine beliefs Medics 4.44 (0.85) 15.10 702.86 <0.001
Non-medics 3.36 (1.06)
Antivaccine beliefs Medics 1.67 (0.89) 15.33 681.59 <0.001
Non-medics 2.76 (1.00)
Risk perception Medics 2.62 (0.88) 0.94 707 0.346
Non-medics 2.68 (0.83)
Perceived gains Medics 4.16 (0.89) 12.55 706 <0.001
Non-medics 3.25 (0.93)
Perceived losses Medics 2.10 (0.82) 14.24 642.14 <0.001
Non-medics 2.99 90.81)
Facilitators’ relevance Medics 3.36 (1.06) 4.30 632.35 <0.001
Non-medics 3.02 (1.02)
Vaccination self-efficacy Medics 3.69 (1.48) 10.10 546.64 <0.001
Non-medics 3.02 (1.02)
A total of 76% of medics and 30% of non-medics declared getting vaccinated in the
previous influenza season (
χ
2
(1)
= 81.14, p< 0.001). However, there were missing data
in the answers to the following question: n= 89 (30%) in medics and n= 235 (58%) in
non-medics. Medics declared more frequent vaccination in the last 5 years than non-medics
(U = 89,888.5, p< 0.001; Me = 4 and Me = 1, respectively) and better access to influenza
Vaccines 2022,10, 723 7 of 13
vaccinations (
U = 19,750.00
,p< 0.001; Me = 1.00 and Me = 4.00, respectively; reversed
scale).
3.2. Verification of the Model
Using SEM allowed us to determine whether there are significant differences between
the groups by comparing the fit indices between the models with certain parameters
constrained to be equal and a model with those same parameters freely estimated (allowed
to differ) across the groups. In the first step, we aimed to confirm that assuming there
were no differences between medics and non-medics would be wrong. To this end, we
tested Model 1 with all parameters allowed and Model 2 with all parameters constrained.
The models differed significantly (p< 0.001). Model 1 (fully unconstrained) obtained a
significantly better fit (
χ2(2216)
= 4291.946, RMSEA = 0.051, CFI = 0.915, SRMR = 0.070)
than the fully constrained Model 2 (
χ2(2266)
= 4536.115, RMSEA = 0.053, CFI = 0.907,
SRMR = 0.112
). These results indicate that the relationships between the determinants of
vaccination intentions are different for medics and non-medics. Thus, the next step in our
analysis was explorative.
3.3. Comparison of the Model in Medics and Non-Medics
In order to compare the fit indices of the entire theoretical model and to explore
the nature of the differences in the model between medics and non-medics, we analysed
the model for each group separately. Our model fitted the data well in both groups, as
summarized in Table 2.
Table 2. Summary of comparison of fit indices of the model between medics and non-medics.
Chi-Square RMSEA Probability
RMSEA 0.05 CFI TLI SRMR
Medics 1844.897 ***, df = 1076 0.049 0.708 0.922 0.916 0.063
Non-medics 2162.701 ***, df = 1076 0.050 0.549 0.926 0.920 0.066
Note: *** = p< 0.001.
3.4. Direct Predictors of Vaccination Intention/Implementation and Self-Efficacy
A table showing the correlation coefficients between the studied variables can be
found in Supplementary Material_2. The results of the SEM conducted separately for each
group are presented in one figure to enable tracking similarities and differences between
groups. As shown in Figure 2, the model explains a noticeable percentage of the Intention
variance as follows: 78% of medics and 56% of non-medics. The medics’ model included a
lower number of significant associations, but they were stronger than in the non-medics’
model. In neither group was age, gender or education a predictor of Intention.
In both groups, the Intention was related to VacSE, and the strength of these connec-
tions was similar (0.45 and 0.48 for non-medics and medics, respectively). The stronger
habit (5-year vaccination) and previous season vaccination were also predictors of Intention,
but low access to vaccines was significant only in non-medics (path coefficient 0.18).
The remaining factors in the model explained the variance of VacSE better in medics
than in non-medics (64% and 49%, respectively). In both groups, the direct predictors of
VacSE were VacGains (stronger in medics) and FacRel. However, only in non-medics was
the role of VacLosses significant (path coefficient
0.54). The high VacLosses are related to
low VacSE.
VacPros directly and positively predicted VacGains and FacRel in both groups, but
more strongly in medics (1.69 versus 0.66 in non-medics). The roles of VacCons and Risk
differed substantially between the groups. In medics, VacCons directly and positively
predicted both FacRel and VacGains (1.63 and 0.77, respectively) and were not related
to VacLosses. In non-medics, they directly and positively predicted VacLosses (path
coefficient 0.53) and were negatively connected with FacRel and VacGains (
0.19 and
0.12,
Vaccines 2022,10, 723 8 of 13
respectively). Only in non-medics did Risk significantly predict FacRel and VacGains (the
association with VacLosses approached significance).
3.5. Indirect Effects
Table 3shows that the groups differed in the number of indirect effects confirmed in
the model. The path from FacRel through VacSE to Intention was significant in both groups.
The same refers to the effects of VacPros on VacSE with VacGains and FacRel as mediators;
however, they were stronger in medics. The paths from VacCons and Risk (through
VacLosses and FacRel) to VacSE were specific only to non-medics. The effect of VacCons on
VacSE through VacGains was significant in medics and only approached significance in non-
medics. However, their directions were the opposite. In medics, VacCons were positively
related to VacGains, which in turn predicted VacSE. In non-medics, the association between
VacCons and VacGains was negative and rather weak.
Table 3. Specific indirect effects between predictors in medics and non-medics (only significant).
Effect (From–To) Group Estimate SE P
FacRel–VacSE–Intention/vac Medics 0.235 0.076 0.002
Non-medics 0.260 0.043 <0.001
VacPros–VacGains–VacSE Medics 1.764 0.453 <0.001
Non-medics 0.239 0.066 <0.001
VacPros–FacRel–VacSE Medics 1.120 0.531 0.035
Non-medics 0.296 0.048 <0.001
VacCons–VacGains–VacSE Medics 0.798 0.390 0.040
Non-medics 0.042 0.022 0.056
VacCons–VacLosses–VacSE Medics – – –
Non-medics 0.286 0.085 0.001
VacCons–FacRel–VacSE Medics – – –
Non-medics 0.107 0.040 0.007
Risk–VacLosses–VacSE Medics – – –
Non-medics 0.083 0.030 0.006
Risk–FacRel–VacSE Medics – – –
Non-medics 0.249 0.052 <0.001
Note: VacPros = pro-vaccination beliefs; VacCons = antivaccination beliefs; Risk = risk perception;
VacGains = outcome expectation of perceived gains; VacLosses = outcome expectation of perceived losses;
FacRel = facilitators’ relevance; Vac SE = vaccination self efficacy; Intention/vac = intention/vaccination.
4. Discussion
The data collected at the beginning of the flu season during the COVID-19 pandemic
allowed us to capture not only the flu vaccination intention but also vaccination implemen-
tation in part of the sample. The results confirmed that medics performed much better
than non-medics and that proportions of readiness to hesitancy were almost inverse in
the groups. This finding confirmed our supposition that the group of medics may be
considered a positive deviant in vaccination performers. By design, the groups differed
in education, but the medics were also older, with a slight predominance of females and
those living in bigger towns. We intentionally focused on medics who, due to their unique
education and experience, are expected to behave in line with medical knowledge. The
problem is that medical education is not a sufficient condition for influenza vaccination, and
there are other factors which may facilitate or hinder this behaviour. Believing that medical
education may promote some process that increase vaccination intention, we wanted to
Vaccines 2022,10, 723 9 of 13
analyse this process and see if it would be possible to activate it among people not trained
professionally in medicine. Thus, we treated the potential bias as an opportunity.
This readiness rate in medics, which is rather high in relation to previous seasons, can
be attributed to the COVID-19-related increase in vaccination motivation [
43
]. However, the
effect confirmed in the general population of the United Kingdom [
44
] was only observed
to a small extent in participating non-medics, which may suggest that medics are more
responsive to external cues for adjusting their health behaviour. Even so, approximately
30% of medics declared vaccination refusal or hesitancy, which clearly shows that the
higher vaccination uptake in medics cannot be attributed solely to their education.
As for the first hypothesis, general pro- and antivaccine beliefs together with risk
perception were related to self-efficacy through influenza vaccination-specific beliefs, but
there were differences between groups. The variance of vaccination self-efficacy in medics
was fully explained only by the paths positively connecting pro- and antivaccine beliefs
with both gains and facilitators.
In non-medics, antivaccine beliefs were related positively to perceived losses and
negatively to gains and facilitators, whereas in medics the first effect did not exist, and
the last ones were positive. One can speculate that these positive relationships in medics
may be the result of unbiased knowledge of vaccines, reasons and likelihood of adverse
events, procedures of testing vaccines and, finally, realistic expectations and evidence-based
critical thinking. This would suggest that there are some benefits to the doubts [
45
] and
that some levels of negative beliefs about vaccines can be conducive to perceiving gains
and facilitators. The other option would be that it is not a matter of the level of negative
beliefs only, but rather the proportion between positive and negative beliefs, similar to
the positivity ratio [
46
,
47
]. For example, the pro-/antivaccine belief ratio in medics was
2.65, whereas in non-medics it was 1.22. Further investigation is needed to explore if this
indicator would be applicable to general vaccination beliefs. The gains/losses ratio can also
be considered [
48
]. Generally, it is believed that gain-framed messages can be more effective
when advocating prevention behaviour such as vaccination [
49
]. On the other hand, some
groups of people for whom loss-framed messages can be more successful, depending on
their level of involvement or knowledge, should not be ignored [
50
]. Probably a more
balanced message emphasising gains but not omitting losses could be effective.
In non-medics, antivaccine beliefs are also negatively related to facilitators. This
connection was even stronger than in medics, which suggests that vaccination campaigns
and interventions for the general public should focus on facilitators such as personalised
reminders, positive modelling by a doctor, a conversation initiated by a doctor, encourage-
ment by supervisors or free access and friendly procedures [23].
Although the groups did not differ in their levels of risk perception, to medics they
were irrelevant. According to HAPA [
18
], risk perception is the determinant of the intention
specific to the motivational stage. This seems to be advantageous mainly initially to put
people on track to developing the motivation to change, but later on, other variables are
more influential in the self-regulation process. In non-medics, risk perception played a
positive role. The above observations would suggest that medics were at a more advanced
stage of behavioural change, where some factors, such as risk perception and vaccination
losses, no longer played a significant role. According to the precaution adoption process
model [
51
], it seems that non-medics predominantly might be still in one of four following
stages: unengaged; undecided; decided not to act; or decided to act. Medics are likely to be
in one of the following stages: decided to act; acting (need resources to act, detailed ‘how-to’
information and self-efficacy); or maintenance. If yes, then the question arises about factors
promoting the transition from one stage to another. Assuming that people at different
stages represent a stage-specific mindset and need to master different tasks, it seems that
intervention tailored to the individual’s stage would be the most appropriate [15]. On the
other hand, it cannot be ruled out that some stages differ in the levels of some predictors
only quantitatively (e.g., each successive stage shows significantly higher self-efficacy or a
stronger intention to perform vaccination).
Vaccines 2022,10, 723 10 of 13
As for the second hypothesis, it was confirmed in both groups. Self-efficacy mediated
the relationships between vaccine beliefs and behavioural intention. In medics, the main
mechanisms regulating vaccination intention/implementation seem to rely on supporting
perceptions of vaccination gains and facilitators’ relevance. The positive connections
existed also in non-medics but were weaker. This indicates that all these vaccine cognitions
should be enhanced because they capture the most relevant factors in the 4C (confidence,
convenience, complacency, calculation) model of vaccine hesitancy [52].
Further differences between groups were observed in direct predictors of vaccination
self-efficacy. Especially intriguing was the path running through perceived losses. The
very tempting supposition would be that as long as the levels of losses were as high as
in non-medics, they were potent enough to decrease self-efficacy, mediating the relation-
ships with antivaccine beliefs. These beliefs, including undesirable vaccine effects, low
seriousness/prevalence of influenza and dishonesty of the vaccine industry [
53
,
54
], were
connected to potentially negative outcomes expected after vaccination, such as social re-
jection, everyday inconveniences and health risks. The problems of uncertainty about the
safety of flu vaccines and mistrust of information provided to the public that dismisses
concerns were previously noticed [45].
The third hypothesis refers to the contextual variables. Apart from vaccine access,
the relationships between vaccination intention/implementation and the most proximal
predictors were similar in both groups. These results are in line with other studies con-
firming the positive effects of vaccination self-efficacy [
21
,
22
,
55
,
56
], previous experiences
(or habit) and last season’s vaccination [
8
,
25
,
57
59
] both in medics and other groups. In
medics, vaccine access turned out to be insignificant, probably due to low variance in this
variable; as a priority group, they had easier access than the general public. Additionally,
having to pay for vaccines is a proven barrier to vaccination [
27
]. In none of the groups
were the included sociodemographic factors significant contributors.
In the light of the above discussion, we may confirm the fourth hypothesis regarding
the differences in the relationships tested in the model between the groups.
Some limitations of the study should be mentioned. Vaccination performance was
not the only difference between the groups, and there might be further contributors to
vaccination intention that should be included in the model. Importantly, the groups
also differed in some sociodemographic characteristics (medics were older, with a slight
predominance of females and those living in bigger towns). These factors may influence
general and specific beliefs about vaccination and distort differences between the study
groups. Importantly, the analysis confirmed the equivalence of scales between the groups,
and these factors were not related to the variables in the model when testing each group
separately. However, the conclusions should be treated with caution.
The numbers in different medical groups were unequal. The self-reported data can be
biassed by social approval, especially as vaccination has recently become a controversial
topic [
60
]. The study design was cross-sectional, which does not allow the hypothesised
causal relationships between predictors to be confirmed. The results, therefore, should
be treated as explorative. The number of potential contributors to vaccination intention
is much higher than those included in our study. We started with a truncated model;
for example, the intention–behaviour gap was not analyzed. Among those who form an
intention, more than half fail to put this intention into practice. This should be analysed in
further studies using the positive deviant approach.
5. Conclusions
It seems that the positive deviant approach adopted in this study allowed us to
broaden our perspective and be better informed on the specificity of the better performer
of influenza vaccination. It was probably the first attempt to use it in the vaccination
context. It seems that providing comprehensive and up-to-date information about the
risks and benefits of vaccination to the public is essential. Reliable knowledge of vaccines
together with clearing up controversies and misconceptions should be complemented by
Vaccines 2022,10, 723 11 of 13
trust-building strategies with information on gains (gains should noticeably outweigh
losses) and facilitators (‘how-to’ and ‘just do’). Among those with high pro-vaccine beliefs,
some level of evidence-based critical beliefs about vaccination can also be beneficial.
In the case of influenza vaccination, which is implemented annually, the individual’s
focus on performance varies over time. Vaccination promotion shortly before the flu season
should target non- or low-intenders and also intenders by the delivery of a balanced scope
of information affecting key vaccination cognitions. Additionally, general pro-vaccine
beliefs, which may act as implicit attitudes, should be created in advance to allow priming
processes after delivering information on flu vaccine availability [61].
Supplementary Materials:
The following supporting information can be downloaded at: https://www.
mdpi.com/article/10.3390/vaccines10050723/s1, supplementary material.
Author Contributions:
Conceptualization, D.W. and U.Z.; methodology, D.W. and U.Z.; software,
D.W. and U.Z., validation, D.W. and U.Z; formal analysis, D.W. and U.Z.; formal analysis, D.W. and
U.Z.; resources, D.W and U.Z.; data curation, D.W.; writing—original draft preparation, D.W. and
U.Z; writing—review and editing, D.W and U.Z.; visualization, D.W.; supervision, D.W.; project
administration, D.W.; funding acquisition, D.W. All authors have read and agreed to the published
version of the manuscript.
Funding:
This research was supported by the Medical University of Warsaw (Grant number
PW/Z/33/8/20(1)).
Institutional Review Board Statement:
The study was conducted in accordance with the Declaration
of Helsinki and approved by the Institutional Review Board (or Ethics Committee) of the Medical
University o Warsaw.
Informed Consent Statement:
Informed consent was obtained from all subjects involved in
the study
.
Data Availability Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
World Health Organization (2019). Recommendations on Influenza Vaccination during the 2019–2020 Winter Season: October 2019 (No.
WHO/EURO: 2019-3718-43477-61074); Regional Office for Europe: Copenhagen, Denmark, 2019.
2.
Mereckiene, J. Seasonal Influenza Vaccination and Antiviral Use in EU/EEA Member States: Overview of Vaccine Recommendations for
2017–2018 and Vaccination Coverage Rates for 2015–2016 and 2016–2017 Influenza Seasons; European Centre for Disease Prevention
and Control: Stockholm, Sweden, 2018. [CrossRef]
3.
Sas, A. Number of Influenza Vaccinations in Poland from 2001 to 2020. Available online: https://www.statista.com/statistics/10
86732/poland-number-of-influenza-vaccinations/#statisticContainer (accessed on 10 January 2022).
4.
Baxter, R.; Taylor, N.; Kellar, I.; Lawton, R. What methods are used to apply positive deviance within healthcare organisations? A
systematic review. BMJ Qual. Saf. 2016,25, 190–201. [CrossRef] [PubMed]
5.
Karafillakis, E.; Dinca, I.; Apfel, F.; Cecconi, S.; W˝urz, A.; Takacs, J.; Suk, J.; Celentano, L.P.; Kramarz, P.; Larson, H.J. Vaccine
hesitancy among healthcare workers in Europe: A qualitative study. Vaccine 2016,34, 5013–5020. [CrossRef] [PubMed]
6.
Le Marechal, M.; Fressard, L.; Agrinier, N.; Verger, P.; Pulcini, C. General practitioners’ perceptions of vaccination controversies:
A French nationwide cross-sectional study. Clin. Microbiol. Infect. 2018,24, 858–864. [CrossRef] [PubMed]
7.
Karafillakis, E.; Larson, H. The paradox of vaccine hesitancy among healthcare professionals. Clin. Microbiol. Infect.
2018
,24,
799–800. [CrossRef]
8.
Schmid, P.; Rauber, D.; Betsch, C.; Lidolt, G.; Denker, M.-L. Barriers of Influenza Vaccination Intention and Behavior—A Systematic
Review of Influenza Vaccine Hesitancy, 2005–2016. PLoS ONE 2017,12, e0170550. [CrossRef]
9.
edrzejek, M.J.; Mastalerz-Migas, A. Influenza vaccination in healthcare workers: Vaccination coverage, determinants, possible
interventions. Med. Pr. 2021,72, 305–319. [CrossRef]
10.
Abraham, C.; Sheeran, P. The health belief model. In Predicting and Changing Health Behaviour: Research and Practice with Social
Cognition Models, 3rd ed.; Conner, M., Norman, P., Eds.; Open University Press: Maidenhead, UK, 2015; pp. 30–69.
11. Ajzen, I. The Theory of Planned Behavior. Organ. Behav. Hum. Decis. Process. 1991,50, 179–211. [CrossRef]
12.
Luszczynska, A.; Schwarzer, R. Social cognitive theory. In Predicting and Changing Health Behaviour: Research and Practice with
Social Cognition Models, 3rd ed.; Conner, M., Norman, P., Eds.; Open University Press: Maidenhead, UK, 2015; pp. 225–251.
13.
Zhang, C.-Q.; Zhang, R.; Schwarzer, R.; Hagger, M.S. A meta-analysis of the health action process approach. Health Psychol.
2019
,
38, 623–637. [CrossRef]
Vaccines 2022,10, 723 12 of 13
14.
Michie, S.; Atkins, L.; West, R. The Behaviour Change Wheel: A Guide to Designing Interventions, 1st ed.; Silverback Publishing:
Sutton, UK, 2014; pp. 1003–1010.
15.
Sniehotta, F.F.; Aunger, R. Stage Models of Behaviour Change. Health Psychology, 2nd ed.French, D.P., Kaptein, A., Vedhara, K.,
Weinman, J., Eds.; Blackwell: Oxford, UK, 2010; (in press)
16.
Brewer, N.T.; Chapman, G.B.; Gibbons, F.X.; Gerrard, M.; McCaul, K.D.; Weinstein, N.D. Meta-analysis of the relationship between
risk perception and health behavior: The example of vaccination. Health Psychol. 2007,26, 136–145. [CrossRef]
17.
Schwarzer, R.; Sniehotta, F.F.; Lippke, S.; Luszczynska, A.; Scholz, U.; Schüz, B.; Wegner, M.; Ziegelmann, J.P. On the Assessment
and Analysis of Variables in the Health Action Process Approach Conducting An Investigation. 2013. Available online:
http://userpage.fu-berlin.de/gesund/hapa_web.pdf (accessed on 10 March 2022).
18.
Schwarzer, R. Health Action Process Approach (HAPA) as a theoretical framework to understand behavior change. Actualidades
En Psicol. 2016,30, 119–130. [CrossRef]
19.
Larson, H.J.; Jarrett, C.; Eckersberger, E.; Smith, D.M.D.; Paterson, P. Understanding Vaccine Hesitancy around Vaccines and
Vaccination from a Global Perspective: A Systematic Review of Published Literature, 2007–2012. Vaccine
2014
,32, 2150–2159.
[CrossRef] [PubMed]
20.
Klimiuk, K.; Czoska, A.; Biernacka, K.; Balwicki, Ł. Vaccine misinformation on social media: Topic-based content and sentiment
analysis of Polish vaccine: Deniers’ comments on Facebook. Hum. Vaccines Immunother.
2021
,17, 2026–2035. [CrossRef] [PubMed]
21.
Godin, G.; Vézina-Im, L.-A.; Naccache, H. Determinants of Influenza Vaccination among Healthcare Workers. Infect. Control Hosp.
Epidemiology 2010,31, 689–693. [CrossRef]
22.
Fall, E.; Izaute, M.; Chakroun-Baggioni, N. How can the health belief model and self-determination theory predict both influenza
vaccination and vaccination intention ? A longitudinal study among university students. Psychol. Health
2018
,33, 746–764.
[CrossRef]
23.
Boey, L.; Bral, C.; Roelants, M.; De Schryver, A.; Godderis, L.; Hoppenbrouwers, K.; Vandermeulen, C. Attitudes, believes,
determinants and organisational barriers behind the low seasonal influenza vaccination uptake in healthcare workers—A
cross-sectional survey. Vaccine 2018,36, 3351–3358. [CrossRef] [PubMed]
24.
Szilagyi, P.G.; Albertin, C.; Casillas, A.; Valderrama, R.; Duru, O.K.; Ong, M.K.; Vangala, S.; Tseng, C.-H.; Rand, C.M.;
Humiston, S.G.; et al. Effect of Patient Portal Reminders Sent by a Health Care System on Influenza Vaccination Rates. JAMA
Intern. Med. 2020,180, 962–970. [CrossRef] [PubMed]
25.
Prematunge, C.; Corace, K.; McCarthy, A.; Nair, R.C.; Pugsley, R.; Garber, G. Factors influencing pandemic influenza vaccination
of healthcare workers—A systematic review. Vaccine 2012,30, 4733–4743. [CrossRef] [PubMed]
26.
Yeung, M.P.S.; Lam, F.L.; Coker, R. Factors associated with the uptake of seasonal influenza vaccination in adults: A systematic
review. J. Public Health 2016,38, 746–753. [CrossRef]
27.
Brondi, L.; Higgins, M.; Gorman, D.; McCormick, D.; McCallum, A.; Fisken, S. Review of the Scientific Literature on Drivers and
Barriers of Seasonal Influenza Coverage in the EU/EEA. Stockh. Eur. Cent. Dis. Prev. Control. 2013. [CrossRef]
28.
Schwarzer, R.; Luszczynska, A. Health Action Process Approach. In Predicting Health Behaviour, 3rd ed.; Conner, M., Norman, P.,
Eds.; McGraw Hill Open University Press: Maidenhead, UK, 2015; pp. 252–278.
29.
Zhou, G.; Gan, Y.; Ke, Q.; Knoll, N.; Lonsdale, C.; Schwarzer, R. Avoiding exposure to air pollution by using filtering facemask
respirators: An application of the health action process approach. Health Psychol. 2016,35, 141–147. [CrossRef]
30.
Ajzen, I. Attitudes and persuasion. In The Oxford Handbook of Personality and Social Psychology; Deaux, K., Snyder, M., Eds.; Oxford
University Press: Oxford, UK, 2012; pp. 367–393.
31.
Albarracín, D.; Chan, M.P.S.; Jiang, D. Attitudes and attitude change: Social and personality considerations about specific and
general patterns of behavior. In The Oxford Handbook of Personality and Social Psychology, 2nd ed.; Oxford University Press: Oxford,
UK, 2018; pp. 439–464.
32.
Ajzen, I. Constructing a TPB Questionnaire: Conceptual and Methodological Considerations. 2002. Available online: https:
//citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.601.956&rep=rep1&type=pdf (accessed on 10 November 2021).
33.
Conner, M. Social cognitions in health behaviour. In Assessment in Health Psychology; Benyamini, Y., Johnston, M.,
Karademas, E.C.
,
Eds.; Hogrefe Publishing: Gottingen, Germany, 2016. [CrossRef]
34. Brown, T.A. Confirmatory Factor Analysis for Applied Research, 2nd ed.; Guilford Press: New York, NY, USA, 2015.
35.
Soper, D.S. A-priori Sample Size Calculator for Structural Equation Models. 2021. Available online: https://www.danielsoper.
com/statcalc (accessed on 10 November 2021).
36. Cohen, J. Statistical Power Analysis for the Behavioral Sciences, 2nd ed.; Lawrence Erlbaum Associates: Hillsdale, NJ, USA, 1988.
37.
Westland, J.C. Lower bounds on sample size in structural equation modeling. Electron. Commer. Res. Appl.
2010
,9, 476–487.
[CrossRef]
38. Hair, J.F.; Black, W.C.; Babin, B.J.; Anderson, R.E. Multivariate Data Analysis, 8th ed.; Cengage Learning: Andover, UK, 2018.
39.
Lubiewska, K.; Głogowska, K. Zastosowanie analizy równowa˙
zno´sci pomiarowej w badaniach psychologicznych. Pol. Forum
Psychol. 2018,23, 330–356. [CrossRef]
40.
Chen, F.F. Sensitivity of Goodness of Fit Indexes to Lack of Measurement Invariance. Struct. Equ. Model. A Multidiscip. J.
2007
,14,
464–504. [CrossRef]
41. Jöreskog, K.G. Simultaneous factor analysis in several populations. Psychometrika 1971,36, 409–426. [CrossRef]
Vaccines 2022,10, 723 13 of 13
42.
Sorbom, D. A general method for studying differences in factor means and factor structures between groups. Br. J. Math. Stat.
Psychol. 1974,27, 229–239. [CrossRef]
43.
Rachiotis, G.; Papagiannis, D.; Malli, F.; Papathanasiou, I.V.; Kotsiou, O.; Fradelos, E.C.; Daniil, Z.; Gourgoulianis, K.I. Determi-
nants of Influenza Vaccination Coverage among Greek Health Care Workers amid COVID-19 Pandemic. Infect. Dis. Rep.
2021
,13,
71. [CrossRef]
44.
Bachtiger, P.; Adamson, A.; Chow, J.-J.; Sisodia, R.; Quint, J.K.; Peters, N.S. The Impact of the COVID-19 Pandemic on the Uptake
of Influenza Vaccine: UK-Wide Observational Study. JMIR Public Health Surveill. 2021,7, e26734. [CrossRef]
45.
Karafillakis, E.; Larson, H.J. The benefit of the doubt or doubts over benefits? A systematic literature review of perceived risks of
vaccines in European populations. Vaccine 2017,35, 4840–4850. [CrossRef]
46.
Fredrickson, B.L.; Losada, M.F. Positive Affect and the Complex Dynamics of Human Flourishing. Am. Psychol.
2005
,60, 678–686.
[CrossRef]
47.
Sabey, C.V.; Charlton, C.; Charlton, S.R. The “Magic” Positive-to-Negative Interaction Ratio: Benefits, Applications, Cautions, and
Recommendations. J. Emot. Behav. Disord. 2018,27, 154–164. [CrossRef]
48.
De Langhe, B.; Puntoni, S. Bang for the Buck: Gain-Loss Ratio as a Driver of Judgment and Choice. Manag. Sci.
2015
,61,
1137–1163. [CrossRef]
49.
Wansink, B.; Pope, L. When do gain-framed health messages work better than fear appeals? Nutr. Rev.
2014
,73, 4–11. [CrossRef]
[PubMed]
50.
Keller, P.A.; Lehmann, D.R. Designing Effective Health Communications: A Meta-Analysis. J. Public Policy Mark.
2008
,27, 117–130.
[CrossRef]
51.
Weinstein, N.D.; Sandman, P.M. A model of the precaution adoption process: Evidence from home radon testing. Health Psychol.
1992,11, 170–180. [CrossRef]
52.
Bhugra, P.; Grandhi, G.R.; Mszar, R.; Satish, P.; Singh, R.; Blaha, M.; Blankstein, R.; Virani, S.S.; Cainzos-Achirica, M.; Nasir, K.
Determinants of Influenza Vaccine Uptake in Patients With Cardiovascular Disease and Strategies for Improvement. J. Am. Heart
Assoc. 2021,10, e019671. [CrossRef]
53.
MacDonald, N.E.; Eskola, J.; Liang, X.; Chaudhuri, M.; Dube, E.; Gellin, B.; Goldstein, S.; Larson, H.; Manzo, M.L.;
Reingold, A.; et al. Vaccine Hesitancy: Definition, Scope and Determinants. Vaccine 2015,33, 4161–4164. [CrossRef]
54.
Ortiz-Sánchez, E.; Velando-Soriano, A.; Pradas-Hernández, L.; Vargas-Román, K.; Gómez-Urquiza, J.L.; La Fuente, G.A.C.-D.;
Albendín-García, L. Analysis of the Anti-Vaccine Movement in Social Networks: A Systematic Review. Int. J. Environ. Res. Public
Health 2020,17, 5394. [CrossRef]
55.
Ling, M.; Kothe, E.J.; Mullan, B. Predicting intention to receive a seasonal influenza vaccination using Protection Motivation
Theory. Soc. Sci. Med. 2019,233, 87–92. [CrossRef]
56.
Real, K.; Kim, S.; Conigliaro, J. Using a validated health promotion tool to improve patient safety and increase health care
personnel influenza vaccination rates. Am. J. Infect. Control. 2013,41, 691–696. [CrossRef]
57.
Hollmeyer, H.G.; Hayden, F.; Poland, G.; Buchholz, U. Influenza vaccination of health care workers in hospitals—A review of
studies on attitudes and predictors. Vaccine 2009,27, 3935–3944. [CrossRef]
58.
Johansen, L.J.; Stenvig, T.; Wey, H. The Decision to Receive Influenza Vaccination Among Nurses in North and South Dakota.
Public Health Nurs. 2012,29, 116–125. [CrossRef] [PubMed]
59.
Lehmann, B.A.; Ruiter, R.A.C.; Wicker, S.; Chapman, G.; Kok, G. Medical students’ attitude towards influenza vaccination. BMC
Infect. Dis. 2015,15, 185. [CrossRef] [PubMed]
60.
Sprengholz, P.; Felgendreff, L.; Böhm, R.; Betsch, C. Vaccination policy reactance: Predictors, consequences, and countermeasures.
2021. Available online: https://psyarxiv.com/98e4t/download?format=pdf (accessed on 10 November 2021).
61.
Cameron, C.D.; Brown-Iannuzzi, J.L.; Payne, B.K. Sequential Priming Measures of Implicit Social Cognition. Pers. Soc. Psychol.
Rev. 2012,16, 330–350. [CrossRef] [PubMed]
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Ending the COVID-19 pandemic will require rapid large-scale uptake of vaccines against the disease. Mandating vaccination is discussed as a suitable strategy to increase uptake. In a series of cross-sectional quota-representative surveys and two preregistered experiments conducted in Germany and the US (total N = 4629), we investigated (i) correlates of individual preferences for mandatory (vs voluntary) COVID-19 vaccination policies; (ii) potential detrimental effects of mandatory policies; and (iii) interventions potentially counteracting them. Results indicate that reactance elicited by mandates can cause detrimental effects, such as decreasing the intention to vaccinate against influenza and adhere to COVID-19 related protective measures.
Article
Full-text available
In the context of the evolving current COVID-19 pandemic, influenza vaccination among the public and health care workers is of particular importance. There are sparse data on the influenza vaccination coverage among health care workers amid COVID-19 pandemic. We aimed to study possible factors associated with influenza vaccination coverage among Greek physicians, dentists, and pharmacists during the flu season 2020–2021. We formulated the hypothesis that COVID-19 pandemic exerted a positive impact on the influenza vaccination coverage of health care workers. An online survey was conducted among the members of the Larissa, (Thessaly, Central Greece) Medical, Dentists, and Pharmacists Associations. The influenza vaccination coverage was 74% (251/340). Acceptance of COVID-19 vaccination coverage was the only factor independently associated with the likelihood of influenza vaccination coverage. In conclusion, we report here a considerable improvement of the influenza vaccination coverage among HCWs and especially among physicians. Acceptance of COVID-19 vaccination was the only predictor of influenza vaccination coverage. This finding provides public health workers and policymakers with useful policy-relevant evidence in order to maintain and even further improve the gains of increased vaccination coverage achieved during the flu season 2020–2021.
Article
Full-text available
Cardiovascular disease (CVD) is the leading cause of death and disability worldwide. Influenza infection is associated with an increased risk of cardiovascular events (myocardial infarction, stroke, and heart failure exacerbation) and mortality, and all‐cause mortality in patients with CVD. Infection with influenza leads to a systemic inflammatory and thrombogenic response in the host body, which further causes destabilization of atherosclerotic plaques. Influenza vaccination has been shown to be protective against cardiovascular and cerebrovascular events in several observational and prospective studies of at‐risk populations. Hence, many international guidelines recommend influenza vaccination for adults of all ages, especially for individuals with high‐risk conditions such as CVD. Despite these long‐standing recommendations, influenza vaccine uptake among US adults with CVD remains suboptimal. Specifically, vaccination uptake is strikingly low among patients aged <65 years, non‐Hispanic Black individuals, those without health insurance, and those with diminished access to healthcare services. Behavioral factors such as perceived vaccine efficacy, vaccine safety, and attitudes towards vaccination play an important role in vaccine acceptance at the individual and community levels. With the ongoing COVID‐19 pandemic, there is a potential threat of a concurrent epidemic with influenza. This would be devastating for vulnerable populations such as adults with CVD, further stressing the need for ensuring adequate influenza vaccination coverage. In this review, we describe a variety of strategies to improve the uptake of influenza vaccination in patients with CVD through improved understanding of key sociodemographic determinants and behaviors that are associated with vaccination, or the lack thereof. We further discuss the potential use of relevant strategies for COVID‐19 vaccine uptake among those with CVD.
Article
Full-text available
Influenza vaccination, as a key element of control activities intended, inter alia, to prevent nosocomial influenza transmission, is recommended each year for all healthcare workers (HCWs). Due to these recommendations and the importance of influenza vaccination among HCWs, a targeted literature review was conducted to assess issues related to vaccination coverage, declared motivators and barriers, as well as interventions to increase vaccination coverage in this professional group. Data obtained from the available publications implies that influenza vaccination rates among WHCs are universally low and vary over time as well as between regions and different types of healthcare professionals (physicians/nurses). One of the main determinants of influenza vaccine uptake is the desire to protect oneself and one's family. On the other hand, practical considerations and misconceptions about the safety and effectiveness of vaccines are the most frequently mentioned preventative causes. There is an urgent need to implement well-organized educational campaigns as this is key to increasing vaccination rates as part of a non-mandatory vaccination strategy. Med Pr. 2021;72(3).
Article
Full-text available
Introduction: Vaccinations are referred to as one of the greatest achievements of modern medicine. However, their effectiveness is also constantly denied by certain groups in society. This results in an ongoing dispute that has been gradually moving online in the last few years due to the development of technology. Our study aimed to utilize social media to identify and analyze vaccine-deniers’ arguments against child vaccinations. Method: All public comments posted to a leading Polish vaccination opponents’ Facebook page posted between 01/05/2019 and 31/07/2019 were collected and analyzed quantitatively in terms of their content according to the modified method developed by Kata (Kata, 2010). Sentiment analysis was also performed. Results: Out of 18,685 comments analyzed, 4,042 contained content covered by the adopted criteria: conspiracy theories (28.2%), misinformation and unreliable premises (19.9%), content related to the safety and effectiveness of vaccinations (14.0%), noncompliance with civil rights (13.2%), own experience (10.9%), morality, religion, and belief (8.5%), and alternative medicine (5.4%). There were also 1,223 pro-vaccine comments, of which 15.2% were offensive, mocking, or non-substantive. Sentiment analysis showed that comments without any arguments as well as those containing statements about alternative medicine or misinformation were more positive and less angry than comments in other topic categories. Conclusions: The large amount of content in the conspiracy theory and misinformation categories may indicate that authors of such comments may be characterized by a lack of trust in the scientific achievements of medicine. These findings should be adequately addressed in vaccination campaigns.
Article
Full-text available
The aim of this study was to analyze social networks’ information about the anti-vaccine movement. A systematic review was performed in PubMed, Scopus, CINAHL and CUIDEN databases. The search equations were: “vaccine AND social network” and “vaccine AND (Facebook[title] OR Twitter[title] OR Instagram[title] OR YouTube[title])”. The final sample was n = 12, including only articles published in the last 10 years, in English or Spanish. Social networks are used by the anti-vaccine groups to disseminate their information. To do this, these groups use different methods, including bots and trolls that generate anti-vaccination messages and spread quickly. In addition, the arguments that they use focus on possible harmful effects and the distrust of pharmaceuticals, promoting the use of social networks as a resource for finding health-related information. The anti-vaccine groups are able to use social networks and their resources to increase their number and do so through controversial arguments, such as the economic benefit of pharmaceuticals or personal stories of children to move the population without using reliable or evidence-based content.
Article
Full-text available
Analysis of measurement invariance is slowly but consistently becoming the standard of quantitative analysis in psychological research. It tests whether a measuring instrument (e.g. a scale) has the same measurement characteristics in groups under the study (e.g. differing in terms of age, culture, or type of research). Establishing the measurement invariance of instruments used in the study allows, at the initial stage of the data analysis, for further testing of the main research hypotheses addressing the relations between constructs or the mean differences. The paper provides an overview of strategies and problems related with the invariance analysis aiming to introduce more widely this method to researchers in Poland. To this end, we first describe levels of measurement invariance and decisions' criteria related with its establishing. We also describe strategies to cope with noninvariance, briefly introducing bayesian methods. Finally, we provide an example guiding step-by-step through the invariance analysis with the use of R environment 'lavaan' package. Summarizing, we highlight the importance of invariance analysis in psychological research postulating that this analysis does not limit researchers but increases the precision of conclusions derived in psychological research. Analiza równoważności skal pomiarowych, czyli testowanie uprawnień do do-konywania porównań i badania relacji pomiędzy wynikami pomiaru w badanych grupach, jest jednym ze sposobów, który pozwala na zdobycie przez badacza (względnej) pewności, że uzyskane przez niego wyniki analiz danych przedstawiają niezafałszowany obraz badanego zjawiska uprawniając do dalszego testowania hipotez głównych o związkach pomiędzy zmiennymi czy różnicach średnich. Inne Adres do korespondencji: Katarzyna Lubiewska,
Article
Background: In the face of the Covid-19 pandemic, the UK National Health Service (NHS) flu vaccination eligibility is extended this season to ~32.4 million (48.8%) of the population. Knowing intended uptake will inform supply and public health messaging to maximise vaccination. Objective: The objective of this study was to measure the impact of the Covid-19 pandemic on acceptance of flu vaccination in the 2020-21 season, specifically focusing on those previously eligible who routinely decline vaccination and the newly eligible. Methods: Intention to receive influenza vaccine in 2020-21 was asked of all registrants of the NHS's largest electronic personal health record by online questionnaire on 31st July 2020. Of those who were either newly or previously eligible but had not previously received influenza vaccination, multivariable logistic regression and network diagrams were used to examine reasons to have or decline vaccination. Results: Among 6,641 respondents, 945 (14.2%) were previously eligible but not vaccinated, of whom 536 (56.7%) intended to receive flu vaccination in 2020/21, as did 466 (68.6%) of the newly eligible. Intention to receive the flu vaccine was associated with increased age, index of multiple deprivation (IMD) quintile, and considering oneself at high risk from Covid-19. Among those eligible but intending not to be vaccinated in 2020/21, 164 (30.2%) gave misinformed reasons. 47 (49.9%) of previously unvaccinated healthcare workers would decline vaccination in 2020/21. Conclusions: In this sample, Covid-19 has increased acceptance of flu vaccination in those previously eligible but unvaccinated and motivates substantial uptake in the newly eligible. This study is essential for informing resource planning and the need for effective messaging campaigns to address negative misconceptions, also necessary for Covid-19 vaccination programmes. Clinicaltrial: Not applicable.
Article
Importance Influenza vaccination rates across the US are low. Because few practices send patient reminders for influenza vaccination, a scalable patient reminder system is needed. Objective To evaluate the effect of patient reminders sent via a health care system’s electronic health record patient portal on influenza vaccination rates. Design, Setting, and Participants This pragmatic, 4-arm randomized clinical trial was performed from October 1, 2018, to March 31, 2019, across the UCLA (University of California, Los Angeles) health care system. A total of 164 205 patients in 52 primary care practices who had used the patient portal within 12 months were included. Interventions Patients due for an influenza vaccine were sent a letter via the patient portal of the health care system reminding them about the importance of influenza vaccination, safety of the vaccine, and morbidity associated with influenza. Patients were randomized within primary care practices to 1 of 4 study groups (no reminder [n = 41 070] vs 1 reminder [n = 41 055], 2 reminders [n = 41 046], or 3 reminders [n = 41 034]). Main Outcomes and Measures The primary outcome was receipt of 1 or more influenza vaccines as documented in the electronic health record, which was supplemented with influenza vaccination data from external sources (eg, pharmacies). Secondary outcomes were influenza vaccination rates among subgroups and influenza vaccinations self-reported by patients in reply to the portal-based query as having been received elsewhere. Results A total of 164 205 patients (mean [SD] age, 46.2 [19.6] years; 95 779 [58.3%] female) were randomly allocated to 1 of the 4 study arms. In the primary analysis across all ages and not including patient self-reported vaccinations in reply to portal reminders, influenza vaccination rates were 37.5% for those receiving no reminders, 38.0% for those receiving 1 reminder (P = .008 vs no reminder), 38.2% for those receiving 2 reminders (P = .03 vs no reminder), and 38.2% for those receiving 3 reminders (P = .02 vs no reminder). In the secondary analysis not including patient self-reported vaccinations, among adults aged 18 to 64 years (vaccination rates: 32.0% in the control group, 32.8% in the 1-reminder group, 32.8% in the 2-reminder group, and 32.8% in the 3-reminder group; P = .001), male patients (vaccination rates: 37.3% vs 38.3%, 38.6%, and 38.8%; P = .001), non-Hispanic patients (vaccination rates: 37.6% vs 38.2%, 38.3%, and 38.2%; P = .004), and those who were not vaccinated in the prior 2 years (vaccination rates: 15.3% vs 15.9%, 16.3%, and 16.1%; P < .001), vaccination rates were higher in the portal reminder groups than in the control group; the findings in these 3 subgroups mirrored the findings in the entire population. When self-reported vaccinations received elsewhere were included, influenza vaccination rates were 1.4 to 2.9 percentage points higher in the portal reminder groups, with a dose-response effect (0 reminders: 15 537 [37.8%]; 1 reminder: 16 097 [39.2%]; 2 reminders: 16 426 [40.0%]; and 3 reminders: 16 714 [40.7%]; P < .001). Conclusions and Relevance Generic patient portal reminders were effective in minimally increasing influenza vaccination rates, but more intensive or more targeted patient motivational strategies appear to be needed. Trial Registration ClinicalTrials.gov Identifier: NCT03666026
Article
Rationale: Seasonal influenza vaccination rates are below the recommended targets, contributing to significant preventable harms. Protection Motivation Theory (PMT), a widely applied model of motivation to respond to threats, may provide some insights into strategies to increase the rate of vaccine uptake. Yet, previous research has omitted some of the proposed predictors of intention when applying this model to vaccination. Objective: The aim of the study is to assess the utility of the PMT in predicting intention to obtain the seasonal influenza vaccine. This study will be the first to examine the role of all six PMT constructs in predicting intention to receive the seasonal influenza vaccine. Method: A cross-sectional study of 547 US residents was conducted using Amazon MTurk. Results: All constructs show significant bivariate correlations in the direction expected from the prior literature. Further examination of the theory within a linear regression model, however, found that perceived costs of vaccinating (i.e., response costs) did not uniquely account for variance in intention. All other components, perceived severity of and susceptibility to influenza, the perceived benefits of not vaccinating (i.e., maladaptive response rewards), the self-efficacy to vaccinate, and the perceived efficacy of vaccinating in preventing influenza (i.e., response efficacy) were unique predictors of intention. Overall, the PMT accounted for 62% of the variance in intention to vaccinate. Conclusions: The study is the first to investigate influenza vaccination using all six theorised predictors of intention from the PMT. The findings highlight the importance of the simultaneous inclusion of all components of the model in assessing their potential utility as targets for intervention. Importantly, the results identify under-utilised constructs in the promotion of vaccine uptake, such as maladaptive response rewards, which should be considered targets for future intervention.