1. Introduction
Annual vaccination is a key public health intervention to reduce the socioeconomic burden of seasonal influenza. The World Health Organization (WHO) recommends [
1] influenza vaccination (IV) for pregnant women, the elderly, children aged 6 months to 5 years, subjects with specific chronic conditions, and healthcare workers (HCWs). Despite these recommendations [
1] and the well-known public health benefits [
2,
3], IV coverage rates are suboptimal in most at-risk groups and jurisdictions [
4,
5].
Italy was the first European country where SARS-CoV-2 spread significantly, with the so-called “first pandemic wave” starting in March 2020 [
6]. Owing to fears of the possible co-circulation of SARS-CoV-2 and influenza viruses, which share several clinical signs and symptoms, some changes were made to Italian IV policy. Specifically, it was recommended that the start of the 2020/21 IV campaign should be brought forward to early October, and the free-of-charge vaccination offer was broadened to children aged 6 months to 6 years and older adults aged 60–64 years [
7]. Some regions introduced compulsory IV for healthcare workers and/or the elderly (especially if institutionalized) [
8,
9,
10]. The first available estimates of 2020/21 IV coverage rates suggest a significant increase in IV uptake among Italian HCWs [
11], the general population, and the elderly [
12].
In May 2020, we conducted a representative cross-sectional survey on attitudes and beliefs concerning seasonal IV [
13]. Briefly, we found that most Italian adults judged IV positively, and that the main determinants of reluctance to receive the 2020/21 seasonal IV were younger age, lower perceived income, and no IV in the previous season. Moreover, about 20% of interviewees declared that, if no COVID-19 pandemic occurred, they would not receive the 2020/21 seasonal IV [
13]. Since that time, several changes have occurred in the epidemiology of both influenza and SARS-CoV-2 infections and associated public health policies. First, the 2020/21 season in Europe saw a 99.8% reduction in influenza virus detections, with only 33 (of 25,606 specimens tested) positive samples reported by sentinel surveillance systems [
14]. In Italy, no influenza virus detections were officially reported [
15]. Second, since December 2020, alpha, beta, gamma, and delta variants of SARS-CoV-2 have been described [
16]; these spread rapidly, owing to their greater transmissibility, becoming dominant in several countries [
17,
18]. Third, in December 2020, the first COVID-19 vaccines were granted conditional approval and a massive immunization campaign began [
19].
These changes prompted us to transform an initial cross-sectional sample into a longitudinal panel in order to monitor public opinion towards (influenza) vaccination at different stages of the ongoing COVID-19 pandemic. In this paper, we describe changes in knowledge, attitudes, and beliefs regarding (influenza) vaccines between the first and second COVID-19 pandemic waves.
4. Discussion
This longitudinal study revealed a significant 1-year increase in respondents’ overall confidence in vaccines and willingness to receive the next seasonal IV. Some socioeconomic and attitudinal factors were independently associated with the propensity to being immunized in the upcoming 2021/22 season. We will now discuss the principal findings in the context of the available Italian and international research and provide suggestions for future IV campaigns.
Both influenza-associated risk perception and preventive behaviors may change over time. This dynamic pattern was, for example, reported during the last A(H1N1)pdm09 pandemic [
23] and the ongoing COVID-19 pandemic [
24]. Indeed, we ascertained a 10% relative increase (from 44.1% in 2020 to 48.6% in 2021 surveys) in people’s intention to be immunized against influenza. It is, however, unclear whether this growth in willingness will translate into increased coverage. On the other hand, in our surveys, we observed a 45.0% relative increase (from 2019/20 to 2020/21) in self-declared IV uptake, which is very close to the officially reported [
12] increase of 41.1% (from 16.8% to 23.7%) in IV coverage in the general population.
Only a few studies on changes in attitudes and beliefs concerning IV in the COVID-19 era are currently available. The Vaccine Confidence Project [
25] reported an increase (from 2018 to 2020) in public trust in vaccines, and in IVs in particular, across most European countries. In Italy, overall vaccine confidence grew from 53% to 60%, while public awareness that IV is important rose from 67.6% to 78.4%. A similar significant increase was seen in other European countries, with the exceptions of Estonia, Romania, Hungary, and the United Kingdom, where no significant changes were recorded. Among other possible reasons, the authors linked this increase to the perceived severity of the COVID-19 pandemic [
25]. An Italian study by Caserotti et al. [
24] reported an increase in willingness to receive IV from the pre-lockdown period (from the end of February to March 8, 2020) to the post-lockdown period (from May 19 to the end of June 2020) (aOR 1.82; 95% CI: 1.16–2.87). By contrast, a six-wave longitudinal study conducted in California between March and August 2020 showed a decreasing trend in intentions to receive IV [
26]. These apparent discrepancies may be explained by several factors, including different survey time periods, and therefore SARS-CoV-2 epidemiology and restrictive public health measures adopted by single jurisdictions. In Italy, the observed increase in intentions to receive IV may also be explained by the “more effective” influenza prevention campaign promoted by the Ministry of Health in 2020 [
27].
According to the effect size observed, the main drivers of the uptake of the next seasonal IV were both previous IV vaccination and COVID-19 vaccination status (or at least intention to be vaccinated against COVID-19). Indeed, IV in the previous season has been systematically shown to determine current IV status in all principal risk groups, including HCWs, the elderly, pregnant women, young children, and subjects with underlying health conditions [
28,
29,
30,
31,
32]. However, we noted that, although IV status in both previous seasons (2019/20 and 2020/21) was a positive predictor of the next IV, the effect size of the latter was about three times higher than that of the former. In an Italian study by Fabiani et al. [
33], elderly subjects vaccinated in the two previous seasons were 8.36 (95% CI: 8.17–8.55) times more likely to receive the 2016/17 IV than those who had not been vaccinated. Indeed, if people have a positive initial experience, most of them will probably seek IV the following year [
29]. It is therefore essential to ensure the continuity of effective IV counseling strategies. In this regard, the role of general practitioners, who are the most trusted source of information on IV, is central.
Unwillingness to receive IV and COVID-19 vaccines may be interrelated. In a UK survey, willingness to undergo COVID-19 vaccination was significantly associated with willingness to receive a 2020/21 IV [
34]. Analogously, among Italian undergraduates, previous IV was associated with greater acceptance of COVID-19 vaccination (aOR 3.81; 95% CI: 1.18–12.27) [
35]. Finally, a meta-analysis of four studies showed that previous IV was a strong predictor (OR 3.17; 95% CI: 1.84–5.46) of COVID-19 vaccine acceptance [
36]. It is therefore likely that effective vaccination counseling on one infection may have indirect positive effects on the other.
Of the structural social determinants analyzed, only increasing age and male sex were associated with a greater intention to receive the 2021/22 IV. That older age is a correlate of IV acceptance has been amply demonstrated by a number of systematic reviews and/or meta-analyses [
28,
29,
30,
31]. By contrast, the role of sex remains to be clarified. Although men tend to report a higher uptake rate than women, the difference becomes non-significant on adjusted analysis [
29]. It is also possible that there is a significant interaction between age and sex; it has been shown that IV uptake in females decreases with increasing age, but increases in males [
37]. In our model, the interaction term was not statistically significant.
In our study, positive attitudes towards IV were associated with greater willingness to receive the next seasonal IV. This finding is in line with those of a systematic review by Schmid et al. [
28], which reported that a negative attitude towards IV was a major barrier to IV uptake in all principal risk groups. Among the various negative attitudes analyzed, lack of confidence in the efficacy of IVs held a prominent place [
28]. A recent US survey [
38] has shown a significantly reduced probability of taking the COVID-19 vaccine if the vaccine effectiveness was 50% (comparison to 70% or 90%), while the difference between a 70% and a 90% protection rate was not statistically significant. Laypeople’s statistical literacy is generally low [
39]. Indeed, Tentori et al. [
40] reported that most people are unaware of the meaning of vaccine effectiveness and confuse this term with the non-incidence rate among vaccinated people. This misinterpretation leads the overall undervaluation of the individual benefits of the vaccine. Moreover, according to the authors, this misunderstanding was aligned with expectations based on misreports in the media [
40]. It has been reported [
41] that the media often report unbalanced messages from the point of view of completeness, transparency, and correctness. We therefore believe that the media should convey their messages through a more balanced reporting and, ideally, adopt shared standards and norms. In our study, the traditional media (e.g., TV or newspapers) showed the largest 1-year increase as a trusted information source. In this regard, interventions such as media training for medical experts and regular meetings that may facilitate communication between experts and journalists may be beneficial [
42].
More generally, it is still unclear whether the expected vaccine effectiveness should be provided to laypeople. Zhao et al. [
43] have stated that “
vaccine effectiveness studies are designed to inform public health decisions rather than for individual decision-making” and “
an individual’s decision to get vaccinated should be primarily informed by their risk of influenza illness and their risk of transmitting influenza to vulnerable people”. We, however, believe that anticipated suboptimal (which varies substantially by season, location, age group, and average 40–60% [
44]) IV effectiveness should be openly disclosed by healthcare providers. In fact, even if the efficacy is only 20% (and the coverage rate is 43%), IV is still able to avert 20.99 million infections, 129,701 hospitalizations, 61,812 deaths, and 2.22 million disability-adjusted life years [
45]. Effective public health communication strategies should therefore provide balanced and laypeople-friendly risk–benefit information on immunization. In turn, effective vaccine promotion campaigns should stress the importance of talking to a healthcare professional about all vaccination aspects, including safety and effectiveness [
46].
For what concerns the main reasons for not having a flu shot, our 2021 survey revealed a substantial 1-year decrease in respondents’ selection of options regarding the presumed low effectiveness of IV. On the other hand, the leading reason for not having a flu shot in the upcoming 2021/22 season was the apparent disappearance of influenza viruses in the previous 2020/21 season [
15]. The epidemiology of influenza and influenza-like illness in the upcoming 2021/22 season is unclear and will probably depend on non-specific COVID-19 pandemic prevention measures, such as social distancing, lockdowns, the wearing of masks, etc. Several scenarios of the evolution of influenza in the COVID-19 pandemic era (e.g., influenza viruses will return, and the same clades will circulate; influenza viruses will return, but some subtypes/lineages/clades will disappear; influenza viruses will return, causing occasional outbreaks) have been proposed [
47]. Moreover, by exerting the so-called “trained immunity” effect (i.e., by boosting the innate immune system), IV may reduce the incidence of some COVID-19-related outcomes. A systematic review and meta-analysis by Wang et al. [
48] recently demonstrated a 14% (aOR 0.86, 95% CI: 0.81–0.91) reduction in the odds of being positive to SARS-CoV-2 in subjects vaccinated against influenza, as compared with non-vaccinated subjects. This argument could also be used to increase IV acceptance. It is clear that influenza (at least type A), being a zoonosis, cannot be eliminated; IV uptake goals [
1] must therefore be pursued.
Over one year, people’s awareness of the existence of different IVs increased. In our opinion, two main factors may have contributed to this increase. First, ongoing controversies over the effectiveness and safety of different COVID-19 vaccines [
49] may have increased people’s willingness to receive one vaccine type rather than another. Second, in 2020, four novel influenza vaccines were authorized and/or commercialized in Italy: quadrivalent egg-based standard-dose adjuvanted, quadrivalent egg-based high-dose, quadrivalent recombinant, and quadrivalent live-attenuated vaccines [
50].
Finally, most interviewees looked favorably on the idea of receiving both IV and COVID-19 shots at the same time and/or a combined influenza/COVID-19 vaccine. From the public health perspective, vaccine co-administration or combined vaccines have several benefits, including fewer missed opportunities to vaccinate, simplified immunization schedules, logistical advantages, and reduced costs [
51,
52]. The first available clinical data [
53] suggest that influenza and COVID-19 vaccines can be co-administered with little interference in terms of immunogenicity and efficacy and only a slight increase in solicited adverse events. By contrast, the rate of unsolicited adverse events was similar among the study arms. The most recent interim clinical considerations for the use of COVID-19 vaccines issued by the Centers for Disease Control and Prevention (CDC) [
54] suggest that SARS-CoV-2 and other vaccines may be co-administered without regard to timing, but each shot should be inoculated in a different injection site. Further large-scale phase III and pharmacovigilance studies on vaccine co-administration (in terms of both immunological inference and safety) are warranted.
Analogously, the first preclinical studies suggest that combined vaccines are promising. A combination of a Matrix-M-adjuvanted quadrivalent nanoparticle IV and NVX-CoV2373 COVID-19 vaccine was immunogenic and efficacious in ferret and hamster models [
55]. Another approach consists of a recombinant influenza type A virus genetic platform that encodes the receptor-binding domain of SARS-CoV-2. This vaccine candidate also proved immunogenic and efficacious against lethal challenge by both viruses [
56].
Our study has some limitations, which should be considered when interpreting the results. First, like all web-based surveys, our study may have been subject to the digital divide bias. However, both probabilistic quota sampling and the longitudinal nature of the survey should mitigate the effects of this bias. Second, self-reported IV uptake was substantially higher than that officially registered. For instance, in this study, self-reported 2019/20 IV uptake (results not shown) was 18.1% (95% CI: 15.5–20.9%), 23.7% (95% CI: 20.9–26.7%), and 45.6% (95% CI: 40.1–51.1%) among subjects aged 18–44, 45–64, and ≥65 years, respectively. The corresponding officially reported [
57] statistics were 3.1%, 9.6%, and 54.6%, respectively. This discordant result is unlikely to have been due to the characteristics of the sample (representative of the adult Italian population) or participation bias (interviewees were not aware of the survey topic beforehand). Possible reasons include the social desirability [
58] and recall (especially in subjects with irregular IV uptake patterns) [
59] biases.