Study setting and design

Qualitative semi-structured individual interviews were employed to better understand older adults’ attitudes, knowledge, and perceptions about influenza disease and vaccination. A mixed team of five male and female undergraduate student interviewers received training that outlined details of the study and provided interviewing techniques with role plays and mock interviews to increase familiarity with the questionnaire and anticipated problems prior to data collection. Participants were invited to the study through telephone calls and emails with senior care facility directors who organized visiting time at their living facilities and local organizations. Individuals where then invited to participate in the study through direct communication in situ. Personal one-on-one interviews were conducted and audio-recorded in private settings at participants’ homes or senior care facilities respectively to maintain confidentiality and to provide a safe space for the participants to share their perceptions without judgment and influence of others. Following the interview protocol approved by Nanyang Technological University’s Institutional Review Board (IRB), all participants were first informed of study objectives as well as the interview procedures and length, and subsequently requested to provide their written informed consent prior to their participation.

Participant selection

Given the study objective to identify salient misperceptions about influenza and vaccination among older adults in Singapore, our inclusion criteria required that participants be Singapore citizens or permanent residents aged 65 and above and be able to converse primarily using the English language. However, Singapore is a polylingual society and while English is widely spoken it is often mixed with terms and phrase usage from other languages. Among older adults in Singapore, the most common second language used is Mandarin Chinese. Thus, all interviewers in our project had to be bilingual and conversant in English and Mandarin Chinese. While interviews were conducted primarily in English, our project team anticipated short usage of Mandarin Chinese by our participants and were trained to conduct member checks [28]. In such cases where a participant used terms or phrases in Mandarin, our interviewers conducted member checks in both English and Mandarin to best match nuances in language use. Participants were compensated with either 20 Singapore Dollars (approximately 15 US Dollars) cash or gift voucher at the end of the interviews. 76 participants who conformed to our inclusion criteria completed the interview, thus providing a robust sample and dataset from to allow for greater descriptive findings and saturation.

Instrument and data collection

We populated questions for our semi-structured interview protocol from reviews of influenza vaccine perception studies [e.g., 29, 30]. Interview topics included health information seeking behavior, influenza knowledge, vaccination history, vaccine perceptions and intentions, and responsibility toward others (refer to S1 Appendix for the interview protocol). The IRB-approved interview protocol provided exact question wording to be used with each participant but allowed for flexibility in the question sequence given the natural flow of discussion. A semi-structured interview format not only ensured that data were exhaustively collected from all participants, but also functioned to minimize interviewer effects. Follow-up questions and probes were also asked during interviews to gather a more elaborated understanding of participants’ thoughts. Interviews were approximately 45 minutes long and were audio-recorded with participants’ consent.

Data management and approach to analysis

In order to facilitate data analysis across this large set of interviews we followed the data management and analytic approach used by Halcomb and Davidson [31]. In this approach, following the completion of each interview, the researchers reflected on their field notes expanding on their initial impressions and documented the major concepts and issues raised by the participants into summarizing field notes using Microsoft Word in order to avoid difficulties inherent to relying solely on audio recording transcriptions [32, 33]. Interviewers were trained to reflect upon the interviews and provided insights on participants’ health misperceptions, personal background, health information seeking, vaccine information access, influenza perceptions, vaccination intentions, vaccine perceptions, social norms, nonverbal behaviors, and any other issues or outcomes from the interview. During this post-interview reflection, the researchers turned back to the audio recording as needed to amend and revise their observations. They also identified and transcribed any direct illustrative quotes from the participants and included them in their field notes while using the raw audio files to corroborate and identify any discrepancies. This approach has been noted to help minimize interviewer bias and improve cross-case comparisons following the documentation of findings [31]. Neither the field notes nor the findings were returned to the participants for comments and corrections.

Following Halcomb and Davidson’s [31] approach, the field notes from all 76 interviews were then compiled into a 295-page compendium used for the primary analysis which employed grounded theory methods to analyze the data [34, 35, 36]. Grounded theory is an inductive process oriented toward building theory that is derived from the data under study [37]. The descriptive and explanatory power of this constructivist approach is advantageous in offering an in-depth understanding of current vaccine perceptions and how those perceptions were formed [38].

Concepts were developed mainly with in vivo coding that reflects the “vernacularity” of responses [39]. The in vivo approach typifies and labels concepts with the verbiage used by participants. This coding process differs from other coding practices such that it uniquely captures participants’ choice of language and nuanced meanings [40] and is particularly relevant to the Singapore context of this study where hybridized dialects have culminated into a version of English known colloquially as “Singlish” that is mixing words from other languages. In our study, in vivo codes provided an understanding of how our participants perceived influenza and vaccination (e.g., “influenza comes from wearing wet clothes” and “influenza vaccine kills the good cells in body”). Subcodes were also identified for some primary-cycle codes to furnish more details (e.g., subcodes for “influenza vaccine can cause side effects” listed examples of vaccine side-effects identified by participants). We involved comparisons of codes first within interviews and subsequently between interviews in order to ensure that all relevant themes were exhaustively catalogued and clustered.

Perceptions about the influenza disease

Perceptions about the influenza vaccine

Perceptions about the influenza disease

Participants’ misunderstanding of disease terms (“influenza” versus “flu”) and the characteristics of influenza demonstrate partial knowledge. Two participants noted the similarities of influenza and cold symptoms but claimed that they have not experienced or been infected with influenza. However, previous studies noted that non-health professionals cannot clearly differentiate influenza from the common cold [43]. Furthermore, in its precise definition, influenza is not a “sickness of the lungs” but rather a respiratory disease that leads to fever and can infect the lungs which may also lead to the development of pneumonia for at-risk individuals.

Participants generally have some rudimentary knowledge of influenza, but also are likely to conflate symptoms with other diseases. The inability to correctly recognize influenza may encourage incorrect prevention and treatment practices among older adults. Further misperceptions about how influenza spreads and how it can be prevented pose significant concerns for this at-risk population as they neither understand the severity of the disease nor have the health knowledge to facilitate adoption of adequate practices to stop the spread of influenza. Therefore, misperceptions about influenza may inadvertently be putting this group at further risk to ancillary diseases and accidents and warrant future engagement regarding influenza transmission and prevention.

Perceptions of good health status and personal immunity as reasons for an individual’s influenza infection were prevalent among participants. These misperceptions are consonant with Sundaram et al.’s [27] findings that healthcare workers perceived influenza “as a routine illness causing temporary inconvenience but no lasting harm” (p. 1999) and with those of some healthcare workers (as found by [44, 45]). This consensus indicates that more communication about the non-relationship between immunity and influenza infection is needed to redress the misperceptions among older adults in Singapore. The prevalence of this “strong immunity and therefore no infection” misconception among both older adults and healthcare workers is concerning, particularly when provider–patient communication may propagate bad health knowledge and impede health promoting behaviors. Influenza communication could therefore emphasize on older adults’ susceptibility and likelihood of developing health complications from influenza to counter this barrier to vaccination.

Further aspects should be taken into account for future communication strategies that aim to reach a target group of 65+ years of age. Older adults face significant challenges concerning memory recall of knowledge-based appeals, and this is exacerbated by age as linear declines in fluency, memory, and perceptual speed are observed as people grow older [46]. Older adults also tend to heuristically process complex information such as those related to health as their cognitive abilities decline with age [46]. Finucane [47] pointed out that older adults are inclined to rely more on emotion-evoking information and suggested that risk information may be portrayed with numeric displays, visuals and in narratives to enhance likelihood of message processing and increase message salience among older audiences. Orel, Spence, and Steele [48] proposed providing factual information to counteract misperceptions and discussing old age vulnerabilities to increase perceived disease susceptibility. Cummings [46] added that written instructions should be accompanied with verbal consultations with healthcare professionals to ensure message comprehension. Message tailoring to older audiences has also been recognized as an effective strategy in increasing information salience and can contribute to reducing information exposure barriers [49]. These recommendations can be incorporated into future communication to improve knowledge about influenza. Sparks and Nussbaum [50] suggested tailoring interpersonal messages to older adults’ personal characteristics and information processing abilities to improve health knowledge. While their recommendation was oriented toward older cancer patients, this advice is not any less useful for encouraging vaccination among older adults now that this group’s primary misperceptions about the influenza disease have been systematically identified in this study.

Perceptions about the influenza vaccine

Our participants also held various misperceptions about the influenza vaccine. Of which, the most common misperception is that the influenza vaccine functions as a travel vaccine. Singapore has an extensive history of disease entering from international travelers, and this seems to influence perceptions that within Singapore, there is little viral spread among residents. News headlines commonly reiterate to get vaccinated when “travelling for your year-end holiday” or before beginning “studies overseas” [13]. While laudable advice, such stories may implicitly contribute to a perpetuated misperception that vaccines including those against influenza should be taken only before traveling and not as routine practice. This suggests that how communication is framed regarding vaccines plays an influential role in shaping perceptions. Therefore, beyond the recommendation that the influenza vaccine should be taken before traveling, vaccination as a routine practice needs to be emphasized to achieve herd immunity and prevent future epidemics.

We noted participants’ confusion of the influenza vaccine with antiviral drugs that are taken to treat influenza infections. This further signifies poor vaccine knowledge among older adults that may explain the low and delayed vaccine uptake (i.e., taking the vaccine only after signs of infection). Poor vaccine knowledge has been recognized as one of the barriers to influenza vaccination [51]. Our study adds that confusion between the functions of vaccines and antivirals is a form of poor vaccine knowledge. Vaccine communication should thus emphasize on the functionality of vaccines, that is, vaccines should be taken before an infection as a preventive measure.

Participants raised further concerns over the safety of the influenza vaccine with regards to its content and periodic manufacturing. This corroborates the SAGE Working Group’s Vaccine Hesitancy Determinants Matrix, in which the introduction of new formulation or new recommendations for existing vaccines was identified to influence hesitancy [52]. Participants’ concerns may be alleviated with communication about vaccine safety (i.e., does not contain germs or zoonotic viruses) from multiple points such as the local government, health organizations, and health practitioners. More comprehensible explanations are necessary to instill knowledge about the mutability of influenza viruses and the need for vaccine manufacturers to produce new vaccines every year to protect users from circulating mutated influenza strains.

Misperceptions about the influenza vaccine are suggested to be exacerbated by inadequate communication about the vaccine. Participants who reported having heard about the influenza vaccine cited doctors as their primary source for vaccine information. Most other participants reported that they have not received any information from their doctors nor had they been recommended the vaccine by healthcare professionals. Even if this is not the actual case—even if these individuals have been informed by healthcare professionals about the influenza vaccine, the message simply is not getting through satisfactorily. This highlights a significant communication gap between doctors and older patients and warrants reinvigorated efforts to adequately engage and ensure that older adults have access and attend to influenza vaccine communication from their healthcare providers. Older patients tend to exhibit less assertive and more paternalistic attitudes toward providers [53] which suggests that older adults may be more willing to comply with direct vaccination recommendations from healthcare professionals than their younger counterparts.

Health engagement initiatives should therefore begin with reinforcement through interpersonal channels via direct communication from healthcare professionals and accompanied by reiterative written instructions and materials that outline the known risk-benefit relationships between influenza vaccination and disease prevention. Many misperceptions may be corrected through adequate direct recommendations from healthcare professionals. Stronger communication efforts to dispel identified vaccine misperceptions and promote vaccine knowledge are warranted.

Limitations and future directions

This study has three limitations that must be acknowledged. First, even though the use of novice interviewers and executive summaries may have been an outcome of limited study resources, our data analysis relied on interviewers’ extensive training, executive summaries and audio playbacks rather than full interview transcripts. This approach introduced the issue of subjective data selection and filtering through interviewers and hence the interpretation of incomplete or reduced qualitative data. Second, the use of nonprobability sampling techniques in recruiting participants limits the generalizability of our qualitative findings. Future replications using stratified sampling techniques are warranted to validate the findings from this present study. Last, language differences are inherent in this study that is conducted in polylingual Singapore where most older adults speak more often in their native language than in English. Despite our interviewers’ language proficiency, interpreting and translating any Mandarin Chinese into English nonetheless pose challenges to the validity of our study findings.

Our study is the first qualitative research in Singapore to identify a typology of its older adults’ misperceptions about the influenza disease and its vaccine. Our findings provide public health practitioners an insight on the complexity of vaccine perceptions among one of the most vulnerable populations. Through this study, we hope that health professionals may be better informed of older adults’ concerns about influenza vaccination and in turn be able to refine their personal communication with this population in encouraging vaccination.