A mixed methods study of seasonal influenza vaccine hesitancy in adults with chronic respiratory conditions

Abstract Background Seasonal influenza vaccination is recommended for patients with chronic respiratory conditions, but uptake is suboptimal. We undertook a comprehensive mixed methods study in order to examine the barriers and enablers to influenza vaccination in patients with chronic respiratory conditions. Methods Mixed methods including a survey (n = 429) which assessed sociodemographics and the psychological factors associated with vaccine uptake (ie confidence, complacency, constraints, calculation and collective responsibility) with binary logistic regression analysis. We also undertook focus groups and interviews (n = 59) to further explore barriers and enablers to uptake using thematic analysis. Results The survey analysis identified that older participants were more likely to accept the vaccine, as were those with higher perceptions of collective responsibility around vaccination, lower levels of complacency and lower levels of constraints. Thematic analysis showed that concerns over vaccine side effects, lack of tailored information and knowledge, and a lack of trust and rapport with healthcare professionals were key barriers. In contrast, the importance of feeling protected, acceptance of being part of an at‐risk group and feeling a reduced sense of vulnerability after vaccination were seen as key enablers. Conclusions Our findings showed that the decision to accept a vaccine against influenza is influenced by multiple sociodemographic and psychological factors. Future interventions should provide clear and transparent information about side effects and be tailored to patients with chronic respiratory conditions. Interactions between patients and their healthcare providers have a particularly important role to play in helping patients address their concerns and feel confident in vaccination.


| BACKG ROU N D
The importance of vaccination and vaccine hesitancy has come into sharp focus due to the COVID-19 pandemic. 1 But the problem of vaccine hesitancy for other vaccines, including the annual seasonal influenza vaccine, is long established. The term "vaccine hesitancy" refers to the "delay in acceptance or refusal of vaccines despite availability of vaccine services". 2 Patients with chronic respiratory conditions are particularly susceptible to serious complications that can arise from influenza, which can lead to hospitalisation. 3 The World Health Organisation (WHO) recommends annual seasonal influenza vaccination for people with longterm medical conditions, including those with chronic respiratory conditions, with a target uptake of 75%. 4 However, vaccination rates across Europe fall well below this. 5 The present study aims to understand the barriers and enablers to influenza vaccination in patients with chronic respiratory conditions.
The reasons for vaccine hesitancy are complex, involving individual influences, contextual influences and vaccine and vaccinationspecific issues. 6 Increasingly, psychological factors are being recognised as providing the best explanations for people not taking up vaccinations. 7 Importantly, these factors are amenable to change through intervention. 6 To date, the research on influenza vaccine hesitancy in patients with chronic respiratory conditions has focussed largely on sociodemographic factors. For example, vaccination rates are higher in patients with comorbidities, while being male, younger in age and a current smoker are all factors that have been associated with lower uptake. 8 However, the research linking influenza vaccine uptake to sociodemographic factors is inconsistent 7 and these factors cannot explain why there has been a rise in vaccine hesitancy over time. Evidence also suggests that psychosocial variables (eg risk perceptions) are better at explaining past vaccination behaviour than demographic, socio-economic and health variables. 9 In a systematic review on the barriers to influenza vaccination, psychological determinants including a lack of confidence (ie trust in the safety and effectiveness of the vaccine and in healthcare professionals and healthcare systems), inconvenience (the ease with which the individual can access the vaccine) and complacency (perceived risk of the illness and necessity of the vaccine), reflected in the "3C" model of vaccine hesitancy 2 were related to uptake across risk groups. 7 More recently, two further "Cs" have been suggested in an extension of the "3C" model, with the "5C" model also including calculation (ie engaging in information searching about the vaccine) and collective responsibility (vaccinating due to a sense of social responsibility). 10 However, to date, no study has undertaken a detailed examination of these psychological barriers to influenza uptake in patients with chronic respiratory conditions. Importantly, knowing the psychological barriers and enablers behind vaccination behaviour can inform the development of evidence-based vaccination policy and practice. 11 A further gap in the literature is that limited attempts have been made to integrate and synthesise qualitative findings on influenza vaccine hesitancy. Specifically in the context of chronic respiratory conditions, no studies have examined the reasons for vaccine hesitancy from a qualitative standpoint. The present study adopted a mixed methods approach, combining quantitative (survey) and qualitative methods (focus groups/interviews). These methods provided complementary insights in order to develop a comprehensive understanding of the psychological barriers and drivers to influenza vaccine uptake in patients with chronic respiratory conditions.

| ME THODS
This exploratory sequential mixed methods study used quantitative (cross-sectional survey) and qualitative (focus groups and interviews) methods. Ethical approval was received from the institutional research ethics committee. Data collection was completed between May and October 2019, before the start of the COVID-19 pandemic.

| Quantitative study
Overall, 429 participants with a chronic respiratory condition completed the survey (participant characteristics are shown in Table 1).
We used convenience sampling with participants recruited to purposive criteria. In order to take part in the survey, participants had to be aged 18-64 and have a chronic respiratory condition, thus meeting the criteria to be offered an annual influenza vaccination free of charge by the National Health Service in the UK. Participants were recruited to the online survey via advertisements on social media and through partner organisations (including the Asthma UK and British Lung Foundation Partnership).

| Key measures
Demographic and disease characteristics: gender, age, type of respiratory condition, severity of condition, educational attainment and socio-economic status (assessed by the Scottish Index of Multiple Deprivation (SIMD)). 12 Vaccination behaviour: Participants were asked whether they had received the influenza vaccination in the previous 12 months (yes/no).
Barriers and enablers to vaccination: The 5C scale assesses psychological antecedents of vaccination and comprises five subscales measuring confidence (eg "I am completely confident that vaccines are safe"), complacency ("vaccination is unnecessary because vaccine-preventable diseases are not common anymore"), constraints ("everyday stress prevents me from getting vaccinated"), calculation ("when I think about getting vaccinated, I weigh the benefits and risks to make the best decision possible") and collective responsibility ("I get vaccinated because I can also protect people with a weaker immune system"). Responses were measured on a sevenpoint scale (1 = strongly disagree, 7 = strongly agree) and scored by calculating the mean for each subscale. 10

| Statistical analysis
Univariate and multivariate binary logistic regression analyses were used to determine the sociodemographic (age, gender, educational attainment and deprivation), health (severity of condition) and psychological factors (confidence, complacency, convenience, calculation and collective responsibility) associated with vaccine uptake (0 = not vaccinated, 1 = vaccinated). All analyses were conducted using IBM SPSS Statistics (version 25) at 5% significant levels.

| Qualitative study
A purposive sampling approach was utilised to recruit adults aged 18-64 years living in Scotland with a chronic respiratory condition. We sought to recruit those participants who over the previous five years had always vaccinated, occasionally vaccinated, and those who never vaccinated in order to understand the full range of vaccine hesitancy. 2 Participants for the qualitative component were recruited using the same methods as the survey participants, but were a distinct group. The focus groups and interviews were conducted by two researchers (KD and AG). In total, 59 participants took part in one of nine focus groups (n = 38) or individual interviews (n = 21). The mean focus group length was 49:27 minutes, and mean interview length was 15:02 minutes. The mean age of participants was 42.8 (SD ± 14.8) years and 70% of participants were female.
Asthma was the most common type of chronic respiratory condition (72%). Of these participants, 24% had not vaccinated in the previous five years, 36% had occasionally vaccinated and 41% had always vaccinated.
Semi-structured focus groups and interviews were conducted were cross-checked by a co-researcher (NC) to offer a further credibility check of the emergent themes. Quotes from the transcripts that captured discrete aspects of each theme were identified.

| Integration of findings
To integrate the findings of the quantitative and qualitative components, we first entered the key results of each component into a table (see Table 4). Drawing on the constant comparative method 14 areas of similarity and difference in the findings were then highlighted by one of the researchers (LW), who generated interpretative statements for the overall meaning of the two contributing studies.
These statements were then checked by the rest of the team to ensure their validity and coherence. severity, 39.6% reported that their condition was mild, 48.5% reported that it was moderate, and 10.7% that it was severe.

| Factors associated with influenza vaccination uptake
Binary logistic regression analysis compared those who had received the influenza vaccine in the previous 12 months (n = 240; 55.9%) with those who had not (n = 189; 44.1%). Univariate analyses showed that there was a significant effect of age, severity of condition, confidence, complacency, constraints and collective responsibility on uptake of the influenza vaccination. There was no effect of gender, educational attainment, deprivation or calculation (see Table 2).
For the multivariate logistic regression, we entered those variables that were significant in the univariate analysis (ie age, severity of condition, confidence, complacency, constraints and collective responsibility). Age, complacency, constraints and collective responsibility remained significantly associated with uptake in the multivariate analysis, but severity of condition and confidence was no longer significant. In the final model, vaccination uptake was associated with being older and having higher levels of collective responsibility, lower levels of complacency and lower perceptions of constraints to vaccination (see Table 3).

| Qualitative study
The inductive thematic analysis highlighted seven themes, four reflecting barriers to vaccination uptake and three reflecting enablers to uptake. The themes that reflected barriers to vaccination uptake were evident among all participants who were occasional or nonvaccinators. The themes that reflected enablers to vaccination were evident among all those who were regular vaccinators.

Feeling cared for and valued
There was a clear sense from participants who vaccinated that they felt their health needs were being catered for. They described the influenza vaccination as easy to get because their GP practice provided a variety of ways for them to obtain it: So we get a letter sent to us from our surgery telling us the date that they're doing the mass vaccination and all the doctors and nurses attend the surgery, Several participants explained that their vaccination was administered by a nurse with whom they were familiar and who showed their concern for them by prompting them to be vaccinated during routine visits. Feelings of being cared for and valued engendered participants' trust in the advice of their HCP: … the nurse is always really nice and jolly and you usually get a good laugh when you go, bit of banter with the nurse, and they always make you feel so TA B L E 4 Integration of the main findings across the quantitative and qualitative studies

Quantitative study Qualitative study Integration and synthesis
Older participants were more likely to get vaccinated The quantitative findings suggest particular targeted efforts may be needed for younger adults with chronic respiratory conditions Having greater confidence in vaccines was associated with getting vaccinated (univariate analysis) The perception that the vaccine causes adverse side effects as well as a lack of trust and rapport in general with HCPs was identified as a barrier The importance of having trust in vaccines and the system that delivers them was identified in both components. Addressing concerns about side effects through transparent communication with healthcare providers may be important Higher levels of complacency about vaccination were associated with not getting vaccinated Acceptance of being at greater risk due to respiratory condition and the belief that influenza represented a significant threat to health' leading to feelings of being protected, was a facilitator to vaccination Both components identified the key role for an individual's perception of risk. Those who perceived influenza to be a significant threat and that they were at greater risk for it were more likely to get vaccinated Participants described the vaccination as giving them "peace of mind", a sense of being "safer" (Oliver, Regular Vaccinator) and "relief" at being protected against influenza (Jill and Anne, Regular Vaccinators).
Nina (Regular Vaccinator) described the vaccination as instilling her with "confidence" in her health. Similarly, another participant said: As long as it continues to help me to live, which is more like, thank God I'm not going to be stuck in my Participants also noted that the sense of protection extended to those around them as well: Yeah, definitely, that sort of help the immunity thing.
I work for a children's palliative care provider, so who we work for is, obviously, it's terminal children, and very complex conditions. So that sort of immunity, from that point of view, I think is really important… (Scott, Regular Vaccinator).

| D ISCUSS I ON
The current study is the first to examine the demographic and psychological barriers and enablers to influenza vaccination in adults with chronic respiratory conditions. Through our mixed methods approach, we examined the psychosocial patterning of vaccine uptake and the psychological processes that influence the decision to vaccinate. The integration of the key findings from the quantitative and qualitative components is shown in Table 4.
These findings demonstrate that having trust in vaccines and HCPs is key. In particular, addressing patients' concerns about side effects through transparent communication may be useful in building trust and confidence. 15 The role of complacency was also important: those individuals who perceived influenza to be a serious illness, and who accepted they were at greater risk were more likely to vaccinate. Participants also noted feeling protected and less vulnerable following vaccination. Practical constraints were also important, with participants emphasising the importance of the information that is made available to them. Many wanted more detailed information that was tailored to their respiratory condition. 16 The survey findings also suggest that having a greater sense of collective responsibility is important, and intervention efforts could aim to foster this by explaining that being vaccinated can protect others. Our findings also indicate that it may be important to target intervention efforts towards younger adults with respiratory illnesses, as they are less likely to get vaccinated.
In terms of intervention development, our findings suggest there are multiple factors that can be targeted in order to improve vaccine uptake. The key role that HCPs can play in shaping their patients' vaccine decision-making was highlighted. 17 Therefore, the development of training and time-efficient strategies for HCPs around how to communicate with vaccine-hesitant patients is warranted. 18,19 Strengths of the study include the mixed methods approach which has allowed us to build a comprehensive picture of the barriers and enablers to influenza vaccine uptake. In addition, our recruitment approach ensured that we recruited participants from across the vaccine hesitancy continuum, including those who were non-vaccinating, those who vaccinated occasionally and those who vaccinated regularly. Limitations include that our findings are limited to participants living in Scotland with a chronic respiratory condition, and so may not be generalisable to other chronic health conditions or national contexts. In addition, as our data were collected prior to the COVID-19 pandemic we do not know what impact the pandemic may have on vaccination beliefs and behaviour in general.

| CON CLUS ION
Our mixed methods study identified a number of sociodemographic and psychological factors that influence influenza vaccination uptake in adults with chronic respiratory conditions. Addressing issues of complacency around the vaccination, building confidence and trust in the vaccine through transparent communication about side effects, overcoming practical constraints including the provision of tailored information and emphasising the importance of collective responsibility can all be utilised in order to increase vaccine uptake.
In addition, targeted interventions for younger adults with chronic respiratory conditions are needed.

ACK N OWLED G EM ENTS
We would like to thank the participants that took part in the study and the Asthma UK and British Lung Foundation Partnership for their support with participant recruitment.

CO N FLI C T O F I NTE R E S T S
No conflict of interest.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1111/irv.12856.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data are not publicly available due to ethical and privacy restrictions.