Elsevier

Vaccine

Volume 33, Issue 22, 21 May 2015, Pages 2620-2628
Vaccine

Phased introduction of a universal childhood influenza vaccination programme in England: population-level factors predicting variation in national uptake during the first year, 2013/14

https://doi.org/10.1016/j.vaccine.2015.03.049Get rights and content

Abstract

Introduction

Through a phased rollout, the UK is implementing annual influenza vaccination for all healthy children aged 2–16 years old. In the first year of the programme in England in 2013/14, all 2–3 year olds were offered influenza vaccine through primary care and a primary school age programme was piloted, mainly through schools, in geographically distinct areas. Equitable delivery is a key aim of the programme; it is unclear if concerns by some religious groups over influenza vaccine content have impacted on uptake.

Methods

At the end of the 2013/14 season, variations in uptake for 2–3 year olds and 4–11 year olds were assessed and stratified by population-level predictors: deprivation, ethnicity, religious beliefs and rurality. GP practice or school level uptake was linearly regressed against these variables to determine potential predictors and changes in uptake, adjusting for significant factors.

Results

Uptake varied considerably by geographic locality for both 2–3 year olds and 4–11 year olds. Lower uptake was seen in increasingly deprived areas, with an adjusted uptake in the most deprived quintile 12% and 8% lower than the least deprived areas by age-group respectively. By ethnicity, the highest non-white population quartile had an adjusted uptake 9% and 14% lower than the lowest non-white quartile by age-group respectively. Uptake also varied according to religious beliefs, with adjusted uptake in 4–11 year olds in the highest Muslim population tertile 8% lower than the lowest Muslim population tertile.

Conclusion

In the first season of the childhood influenza vaccination programme, uptake was not uniform across the country, with deprivation and ethnicity both predictors of low uptake in pre-school and primary school age children, and religious beliefs also an important factor, particularly the latter group. With the continued rollout of the programme, these population-level factors should be addressed to achieve sustained successful uptake, along with assessment of contribution of individual and household-level factors.

Introduction

Until recently, the United Kingdom (UK) has employed a selective influenza immunisation programme, with vaccination targeted at those individuals who are more likely to develop severe disease following infection together with their carers to reduce the risk of transmission. This included 65+ year olds, under 64 year olds in a pre-defined clinical risk group, pregnant women and frontline healthcare workers. However, a considerable burden of disease due to influenza continued to occur in the population in both vaccinated and non-vaccinated groups [1], [2], [3] and evidence is accumulating of the key role that children play in driving influenza transmission each winter [4]. Mathematical modelling work has demonstrated that vaccinating healthy children could potentially provide both direct protection to the targeted groups and indirect protection to other groups by reducing transmission in the population [5], [6]. On this basis, the UK Departments of Health recommended extending the programme to all children aged 2–16 years of age [7], [8].

This programme extension will primarily utilise a newly licensed intra-nasally administered live attenuated influenza vaccine (LAIV), with a phased rollout over a number of seasons. In England in 2013/14, all children aged 2–3 years were offered LAIV through primary care, together with primary school aged children aged 4–11 years old in seven geographically distinct pilots where vaccine was delivered predominantly in school settings [7]. Uptake of influenza vaccine through primary care reached 41.4% in 2–3 year olds [9], 42.6% in 2 year olds and 39.5% in 3 year olds, with a higher uptake reported in at-risk groups (56.5%) relative to healthy children (40.6%). An overall uptake of 52.5% was reached in 4–11 years [10], increasing to 56.3% when considering the six pilot sites where the vaccine was delivered through schools (LAIV was delivered through community pharmacies in the other pilot area).

While these figures appear encouraging in the first year of the programme, there was uncertainty about acceptability and equitable delivery - key requirements of a national vaccination programme. Evidence suggests several factors can have an impact on influenza vaccine uptake. Deprived areas have been shown to have lower uptake across age groups [11], [12], [13] and uptake has also been associated with ethnic diversity, with lower uptake observed in communities with a higher proportion of non-White residents [12]. Lower uptake has also been seen in urban relative to rural areas, though this was largely explained by other factors [13]. Specific concerns were raised during 2013/14, particularly amongst Muslim and Jewish religious groups, regarding the porcine origin gelatin component of LAIV [14]. Despite support for use of the vaccine from groups such as the Muslim Council, the issue received considerable media attention. It is currently unclear what the effect of this might have been on the influenza vaccine programme in 2–3 and 4–11 year olds as religious beliefs have previously been associated with increasing likelihood of under-immunisation for several vaccines [15], [16], [17], [18].

While the majority of studies have assessed predictors of uptake at the individual level, uptake will also be influenced by factors at the level of household and community [19], [20]. Although more limited in the causal inferences that can be drawn, ecological analysis can use routinely collected data to help rapidly identify potential factors associated with low uptake, which can inform subsequent rollout of a vaccination programme. Therefore the key aims of this manuscript were to:

  • Assess the variation in influenza vaccine uptake across England in 2–3 year olds in primary care and 4–11 year olds in primary schools in the first year of the programme in 2013/14 by key population-level factors such as deprivation, ethnicity and religious beliefs.

  • Identify at an ecological level predictors of vaccine uptake in 2–3 year olds and 4–11 year olds to ascertain if inequalities are apparent.

Section snippets

Uptake

The collection of influenza vaccine uptake data has previously been described [12], [13]. Uptake in 2–3 year olds, available stratified by year of age and by predefined clinical risk group, was collected at GP practice level across England through the Public Health England (PHE)-commissioned website, ImmForm [21]. This collection process is mandatory, and predominantly undertaken via automated extraction (>80% of practices). Uptake was calculated by dividing the number of GP-registered 2–3 year

Descriptive analysis

Influenza vaccine uptake in 2–3 year olds was reported by 7175 GP practices in England (>99%), all of which were matched to a lsoa.

Variation in uptake was seen across England (Fig. 1), ranging from 18.8% in a CCG in PHE London region to 60.5% in a CCG in PHE Central region. Low uptake was apparent across London (consistent within 2–3 year olds (Fig. A1)), with an overall uptake of 31.0% compared to 42.3% in Central, 43.8% in North and 44.0% in the South (Table 1). For uptake in 4–11 year olds,

Discussion

Subnationally, considerable variation in influenza vaccine uptake was seen for both pre-school and primary school age programmes, with a range in CCG-level uptake from 19% to 61% for 2–3 year olds and 46% to 64% at the pilot level for 4–11 year olds. While PHE region-level uptake in 2–3 year olds was considerably lower in London, this difference was not significant when controlling for population-level factors, with adjusted uptake only slightly lower in the South. In 4–11 year old pilot areas,

Funding

This work was supported by Public Health England where the lead author and corresponding author are employees. This work was supported through core surveillance funding from the Department of Health. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Conflict of interest

None declared.

Acknowledgements

The authors would like to acknowledge all the PHE and NHS staff in each of the seven pilot sites (Bury, Cumbria, Gateshead, Havering, Leicester City, East Leicestershire and Rutland, Newham and South East Essex) who were responsible for planning and delivering the LAIV primary school-age programme. We are grateful to Odette Eugenio and the ImmForm Team for managing the collation of the GP vaccine uptake data in 2–3 year olds.

References (33)

  • Department of Health

    The flu immunisation programme 2013/14 – extension to children

    (2013)
  • Department of Health (DoH)

    Joint Committee on Vaccination and Immunisation. Minute of the meeting held on Friday 13 April 2012

    (2012)
  • Public Health England

    Influenza immunisation uptake in GP patient groups. Data Collection Survey 2013/14

    (2013)
  • R.G. Pebody et al.

    Uptake and impact of a new live attenuated influenza vaccine programme in England: early results of a pilot in primary school-age children, 2013/14 influenza season

    Euro Surveill

    (2014)
  • K.M. Charland et al.

    Clinic accessibility and clinic- level predictors of the geographic variation in 2009 pandemic influenza vaccine coverage in Montreal, Canada

    Influenza Other Respir Viruses

    (2014)
  • N.A. Otieno et al.

    Demographic, socio-economic and geographic determinants of seasonal influenza vaccine uptake in rural western Kenya, 2011

    Vaccine

    (2014)
  • Cited by (21)

    • Implementation of the United Kingdom's childhood influenza national vaccination programme: A review of clinical impact and lessons learned over six influenza seasons

      2020, Vaccine
      Citation Excerpt :

      School-based vaccination generally resulted in higher vaccination uptake than GP practices (Fig. 4). Analysis of predictors of uptake from the 2013/14 season suggested that ethnicity, religious beliefs, and deprivation (assessed in terms of income, employment, health, education, crime, service access, and living environment) predicted low uptake in preschool- and primary school-aged children [53]. Other studies in England, Wales, and Scotland have found that uptake as part of the UK’s influenza childhood NVP is low among children from ethnic minorities, those living in areas of higher deprivation, and those from larger families [54,55].

    • Population-level factors predicting variation in influenza vaccine uptake among adults and young children in England, 2015/16 and 2016/17

      2018, Vaccine
      Citation Excerpt :

      Areas with high proportion Muslim populations showed a significantly lower adjusted vaccine uptake among children aged 2–4 years and primary school aged children during the 2016/17 influenza season [10]. This effect was not observed during the 2015/16 season for 2–4 year olds nor was it observed 2013/14 season among children aged 2–3 years old [6]. That being said, the effect was observed among children aged 4–11 years old in pilot areas that year [6], indicating that vaccine uptake has varied among Muslim populations over time.

    • Sociodemographic predictors of variation in coverage of the national shingles vaccination programme in England, 2014/15

      2017, Vaccine
      Citation Excerpt :

      After adjusting for ethnicity and LT, GP practice deprivation was an indicator of overall vaccine coverage with stepwise reductions in coverage as deprivation increased. Similar findings have been reported elsewhere [19,21,27] including for the prenatal pertussis and rotavirus immunisation programmes in England [Lisa Byrne, personal communication]. Despite shingles vaccine being offered routinely, the deprivation gradient suggests that some barriers remain to delivering an equitable programme across areas with high and low deprivation.

    View all citing articles on Scopus
    View full text