Exploring strategies to promote influenza vaccination of children with medical comorbidities: the perceptions and practices of hospital healthcare workers

In Australia, high-risk children (HRC) including those with medical comorbidities, aged 6 months and greater are recommended to receive the influenza vaccine each year [1]. The vaccine is funded under the National Immunisation Program and has been freely available to this group since 2010 [2]. Nevertheless, vaccine coverage is suboptimal. Estimates among hospitalised HRC range from 5 to 26.9% [3‐5], with a higher Fig. (41%) reported in a survey conducted in two Sydney hospitals [6]. A recent survey in 2017 across three large paediatric hospitals (located in Western Australia, Victoria and Queensland) documented the uptake at 52% for children with medical comorbidities, however this study was conducted during an unusually high burden influenza season [7].

Health care workers (HCWs) play a pivotal role in vaccine uptake. It is well documented that an HCW recommendation is a strong predictor of influenza vaccination in HRC [6, 8‐11]. Physicians are a trusted source of information whom parents rely upon to initiate vaccine conversations [12, 13]. However, recently it was reported that a recommendation from a hospital-based physician is a stronger predictor of influenza vaccination for Australian HRC, compared to a recommendation from a primary care physician [7]. The study also identified that hospital-based physicians were the caregivers’ most commonly reported source of trusted vaccination information. Parents may perceive that their child’s specialist has more detailed knowledge regarding the child’s condition and their treatment regiments. The problem is however that hospital HCWs may lack knowledge regarding the current guidelines/recommendations regarding influenza vaccination [14]. In a survey of pediatricians in two Sydney hospitals, only half (56%) had a correct understanding of the vaccine recommendation for HRC. Furthermore, approximately 14% believed there were restrictions for children under 5 [15], which is significant considering the large morbidity in this age-group [16]. HCWs have also reported difficulties identifying HRC during consultations [17], as individual physicians may be less familiar with the medical history of a high-risk child, especially if he/she sees multiple providers [18].

HRC have frequent contact with hospitals and secondary care teams [19, 20]. Nevertheless, missed opportunities to vaccinate occur in hospitals; survey data from two Sydney hospitals, for instance, indicated that unvaccinated HRC attended a median of three outpatient visits during the influenza season in which they did not receive the vaccine [6]. Children often obtain the same immunisation status over successive seasons, with those experiencing a missed opportunity to vaccinate in one season being most likely to experience another in the next. Thus, it has been suggested that missed opportunities to discuss or provide the vaccine are not isolated events, but representative of systematic failings in the practices of HCWs [21]. Before we can design new interventions to improve uptake in this population, it is important to understand the current landscape around the promotion and delivery. We conducted a qualitative study involving HCWs at one large Sydney paediatric hospital, to understand what is happening around influenza vaccination of HRC.

A qualitative study was conducted involving in-depth interviews with a sample of hospital staff at one Sydney tertiary paediatric hospital. A purposeful, non-probability sampling method was used to recruit HCWs that were: involved in the clinical care of HRC and/or involved in the promotion or delivery of the influenza vaccine to HRC. Recruitment occurred from April – July 2018. Ethics approval was obtained from the Sydney Children’s Hospital Network (LNR/18/SCHN/24) and written informed consent was received from all participants.

In 2018, free influenza vaccination was available in seven of eight Australian jurisdictions to all children ≤5 years of age. As outlined in the introduction, free influenza vaccines had been available for children with certain HRCs as part of the National Immunisation program since 2010. In Australia, the majority of childhood immunisation is provided in the primary care setting, and a smaller proportion delivered via school-based programs. In some of the larger Australian paediatric hospitals, Immunisation Specialist Services or clinics (with dedicated staff) have been operating to provide better access to children with chronic medical conditions. During the influenza season, many of these clinics provide a drop-in service to allow patients and their families member (at a small cost, if they are not eligible) to receive the influenza vaccine. While the study site is a paediatric public (teaching) hospital with an emergency department, it does not currently have a dedicated immunisation clinic. In this hospital, influenza vaccination is provided opportunistically by staff in the outpatient setting.

The researchers liaised with a staff member who worked with HRC at the hospital to identify key stakeholders who met the inclusion criteria. This included department heads, pediatricians, clinical nurse consultants, and nurses. Department heads were contacted firstly and then were asked to pass on the invitation to any staff members they deemed appropriate for the study. Targeted recruitment was also used to ensure that we captured members of the nursing staff that were known to be involved with the current vaccine delivery processes onto the study. HCWs who consented to the study were contacted via email to organise a face-to-face or telephone interview at a convenient time. Participants were also asked to recommend other HCWs who would be suitable for the study (i.e. snowball recruitment). Due to the recruitment process, we were unable to capture any information on staff who declined to participate.

Semi-structured interviews (15–60-min duration) were conducted by researchers who were independent of the hospital. An interview guide (Additional file 1) containing topics of interest was used to guide discussions. Questions asked pertained to the participant’s current practices, the perceived attitudes of parents, challenges in vaccine delivery, and potential interventions that may increase vaccine uptake. Open-ended questions, paraphrasing, and additional questions were used in a flexible interview approach to allow the exploration of a wide range of ideas. Interviews were audio-recorded with the participant’s permission. At the end of each interview, the main ideas expressed were summarised back to the participant to check the researcher’s understanding. No field notes were collected, repeat interviews were not undertaken, nor were transcripts returned to participants. Most interviews were conducted face to face in the hospital by Holly Seale, Senior Lecturer (PhD, MPH, BSc) who has previously undertaken and published qualitative research focused on influenza vaccination. Vanesa Ma, a medical student, received training to undertake in-depth interviews who sat in on the interviews and undertook three of them. There were no established relationships between the research team and the participants prior to the study. At the end of each interview, a preliminary analysis of the data was undertaken by making note of concepts, ideas and potential themes. This process assisted with identifying further areas for exploration in subsequent interviews and assisted with identifying further departments/staff members to contact and recruit. The analysis continued throughout the data collection period, as the team constantly looked at and discussed the ideas arising from the data. All interviews were audio recorded and professionally transcribed verbatim. The researchers used NVivo 12 to code the data [22]. In analysing the data we followed the steps proposed by Braun and Clarke [23], i.e. transcription of interviews, familiarization by reading and re-reading and looking for meanings and patterns for coding in the data, generating initial codes, generating and reviewing themes, and finalizing the analysis. A pre-determined theoretical/conceptual coding framework was not utilised during the analysis process. Rather, the codes arose based on the participants language, perceptions and sentiments. Feedback on the themes was not sought from participants, as we tried to reduce the research burden on the hospital staff involved.

Seventeen HCWs (9 doctors, 8 nurses) were interviewed between April and July 2018. Participants had worked in healthcare for an average of 28 years (range: 10–56 years) and the majority (14/17) had received the influenza vaccine in 2017. The study results are presented below according to the themes identified:

Given the suboptimal influenza vaccine coverage for children with chronic and underlying medical conditions, this study was undertaken to examine the current landscape around the promotion and delivery of seasonal influenza vaccines at a major paediatric hospital in Sydney. Not surprising that we found that staff members who were responsible for caring for children hospitalised with influenza, were those who were more likely to consider conversations about vaccination as an ‘ingrained’ element of their consultations. They also seemed more willing to have multiple follow-up conversations with parents and had systems in place to get the children vaccinated onsite. This is important as higher coverage has been seen when clinicians continue to nudge parents in spite of resistance [24].

Not all staff members felt that it was the hospital’s responsibility to promote and deliver the influenza vaccination. For some, they preferred to designate responsibility with the primary care physician. Competing priorities during the hospital consultation, lack of resources to deliver vaccination, coupled with the fact that not every HRC will have a hospital consultation during the influenza season were just some of the reasons purported in support of primary care provision. The problem here is threefold: firstly, as previously alluded to, parents are placing higher value on a recommendation from a specialist/hospital physician compared to a primary care physician [7]. Supporting this is the findings from recent Australian studies that found that parents who receive a recommendation from their paediatrician or specialist are 4–16 times more likely to immunise their child [7, 25]. Secondly, general practitioners report being unsure of their role in the management of HRC [26], locating the responsibility of vaccine conversations with the child’s specialist [27]. Lastly, not all children attending the hospital setting may attend primary care regularly or have a regular GP. Merckx et al. [28] for instance, found that almost one-fifth of children immunised in a hospital clinic had no primary care doctor. Even if a family does see a GP, by requiring a child to receive the vaccine in the primary care setting and not immediately onsite, we are adding to the burden of healthcare visits. While these findings support the need for stronger communication between the primary and tertiary care sector and the need to support training of primary care providers to improve their confidence in communicating about vaccination to HRC, they also support the need for hospitals and the staff members to reflect on the key role they could play in supporting improvements in coverage in this at-risk population.

Overwhelmingly, participants spoke about the fact that they did not receive updated information about the influenza vaccine and about recommendations for HRC. This information may have been available but due to conflicting priorities, it had not come onto the staff members radar. Confusion surrounding where to access information within the hospital’s intranet may have contributed to the knowledge gaps in our participants. This is not the first time this finding has been reported. Previously, Philips et al. identified that Australian medical staff are unaware of immunisation resources, including telephone advice lines and government information, highlighting the need for greater promotion of existing tools [29]. Given that HCWs often lack the time and capacity to conduct their own research [30], they require easy access to educational material and resources [31]. The problem is around how to effectively deliver these materials to staff. Posters are often not seen on walls, while printed materials may get sidelined. In this setting, having in-service sessions which including tailored information about the current recommendations, delivered at the department level may assist, in addition to information sessions delivered through-out the influenza season at the hospital level. Having a vaccination champion within each department may also assist with supporting staff to promote and/or deliver vaccine on-site. To support parents to speak up about the vaccine, there is still value in having promotion materials (in a range of languages) at the healthcare facility.

Competing priorities has previously been identified as a barrier to lengthy vaccine conversations by specialists in the United States [31]. This issue was also a common denominator amongst our participant especially amongst HCWs who care for children with highly complex and multiple medical conditions. Competing priorities is not an issue unique to the tertiary sector but also clearly exists for primary care providers [32] as well. In both settings, there is a need for further work to explore how to support providers to value and prioritise discussions about vaccination during their consultations.

One suggestion to assist with the afore mentioned issue is to introduce interventions which targeted parents, such as education and reminder/recall systems. As there is limited time for patient education during consultations, educational tools (e.g. fact sheets) may assist HCWs in communicating the need for vaccination [31]. Martin et al. [33], for instance, employed an asthma education tool during consultations to increase vaccination rates from 37 to 92% among children with persistent asthma. While this study involved only a subset of HRC and is limited by its small sample size, it suggests a strong need to engage parents in vaccine conversations. Conversely, the use of reminder/recall systems in primary care and private practices has achieved only moderate effects [34], with Szilagyi et al. [35] suggesting such methods have a “ceiling” coverage of approximately 30%. Multi-component interventions (i.e. “bundles”) that target both HCWs and parents are thus likely required to achieve optimal coverage within this hospital.

Opportunistic immunisation through the hospital has been shown to increase uptake. Pappano et al. [36], for instance, found higher coverage (57%) among at-risk children who were offered the influenza vaccine in a paediatric emergency department compared to those who were given education on where to obtain the vaccine (36%). Furthermore, while participants in this study were reluctant to vaccinate inpatients, on-ward vaccination programs have been shown to be feasible when there is collaboration with the inpatient clinical team [37]. The provision of influenza immunisation through dedicated vaccination clinics has also been trialled in several Australian hospitals, with a subsequent increase in the number of vaccinations administered [38]. Such interventions, however, often require external funding and the employment of additional staff [28, 39]. A lack of support from other HCWs and minimal incorporation into the local health setting are further barriers to their implementation [40]. Additional resources would be required before increased on-site vaccination is feasible and sustainable in this hospital.

Our study has several limitations. Firstly, the study had a small sample size which limits the transferability of our results to other hospital settings. As we did not interview a clinician from all departments with HRC, we cannot rule out the possibility of missing additional themes or barriers. Emails inviting staff members to participate were sent out via Department Heads and so we were unable to document the number of staff members that declined and the reason. Furthermore, participants in our study had generally positive attitudes towards the influenza vaccine. HCWs who are highly disengaged with the value of influenza immunisation are unlikely to volunteer to be involved. In a recent survey of Sydney pediatricians, less than half (48%) considered the influenza vaccine useful [15]. Thus, ‘negative cases’ warrant further investigation to better characterise the perspectives of HCWs who may be reluctant to promote or provide the vaccine. Lastly, we did not undertake interviews with parents and so the comments around parental acceptance/concerns about the influenza vaccination are based on the perceptions of the HCWs. Despite this, we feel that the findings of this study are relevant to other hospital sites that don’t currently have dedicated immunisation clinics but who could consider enhancing their role in vaccine promotion/delivery.

HCWs play a key role in influenza vaccine uptake in HRC. Active promotion by HCWs and on-site delivery of the vaccine may facilitate immunisation in the hospital setting. Conversely, HCWs may refrain from vaccine conversations due to a lack of knowledge and the presence of competing priorities. Staff members must be considered when designing hospital-based interventions to ensure they are appropriate for the local health setting. A ‘bundle approach’ that targets HCWs and parents is likely required to achieve optimal vaccine coverage at this hospital. This would involve: 1) educating HCWs on the current recommendations for HRC; 2) motivating HCWs to discuss the influenza vaccine in all consultations; 3) increased engagement with parents, and 4) additional resources to allow more streamlined vaccine delivery (e.g. additional nurses).