Effect of mass paediatric influenza vaccination on existing influenza vaccination programmes in England and Wales: a modelling and cost-effectiveness analysis

Abstract

Background: In 2013 England and Wales began to fund a live attenuated influenza vaccine programme for individuals aged 2-16 years. Mathematical modelling predicts substantial beneficial herd effects for the entire population as a result of reduced influenza transmission. With a decreased influenza-associated disease burden, existing immunisation programmes might be less cost-effective. The aim of this study was to assess the epidemiological effect and cost-effectiveness of the existing elderly and risk group vaccination programme under the new policy of mass paediatric vaccination in England.

Methods: For this cost-effectiveness analysis, we used a transmission model of seasonal influenza calibrated to 14 seasons of weekly consultation and virology data in England and Wales. We combined this model with an economic evaluation to calculate the incremental cost-effectiveness ratios, measured in cost per quality-adjusted life-years (QALY) gained.

Findings: Our results suggest that well timed administration of paediatric vaccination would reduce the number of low-risk elderly influenza cases to a greater extent than would vaccination of the low-risk elderly themselves if the elderly uptake is achieved more slowly. Although high-risk vaccination remains cost-effective, substantial uncertainty exists as to whether low-risk elderly vaccination remains cost-effective, driven by the choice of cost-effectiveness threshold. Under base case assumptions and a cost-effectiveness threshold of £15 000 per QALY, the low-risk elderly seasonal vaccination programme will cease to be cost-effective with a mean incremental cost-effectiveness ratio of £22 000 per QALY and a probability of cost-effectiveness of 20%. However, under a £30 000 per QALY threshold, the programme will remain cost-effective with 83% probability.

Interpretation: With the likely move to decreased cost-effectiveness thresholds, reassessment of existing risk group-based vaccine programme cost-effectiveness in the presence of the paediatric vaccination programme is needed.

Funding: National Institute for Health Research, the Medical Research Council.

Figure 1

Effect of low-risk elderly and high-risk vaccination programmes in the presence of paediatric vaccination administered at different speeds through the influenza season Preschool are children aged 2–4 years, primary are children aged 5–10 years, and secondary are adolescents aged 11–16 years. The paediatric vaccination uptake speeds are associated with the accumulation of the fixed paediatric vaccine coverage across October, November, December, and January.

Figure 2

Cost-effectiveness of low-risk elderly and high-risk vaccination programmes in the presence of paediatric vaccination Preschool are children aged 2–4 years, primary are children aged 5–10 years, and secondary are adolescents aged 11–16 years. (A) All paediatric vaccination coverage is administered at speeds consistent with those reported during the full rollout of preschool or the pilot rollout of school-age children. (B) All paediatric vaccination coverage is administered by the end of October (fast uptake). Cost-effectiveness regions are coloured for readability: more than £30 000 per QALY (orange; not cost-effective), £20 000–30 000 per QALY (pink; cost-effective under current protocol), £15 000–20 000 per QALY (green; very cost-effective under current protocol), and less than £15 000 per QALY (blue; cost-effective under proposed protocol). The dark grey area corresponds to the incremental cost-effectiveness of a high-risk vaccine programme in the presence of the respective paediatric vaccine programme. The white areas correspond to the incremental cost-effectiveness of an elderly vaccine programme in the presence of the respective paediatric programme. Each grey and white probability distribution area is scaled to have a fixed maximum height, whereas each distribution represents an area of one unit. The mean (solid), 75% (dashed), and 90% (dotted) quantiles are shown to indicate in which cost-effectiveness region the mean of the distribution, 75%, or 90% of its simulations lie. QALY=quality-adjusted life-year.

Figure 3

Effect of the whole-season direct effectiveness of live attenuated influenza vaccine on the probability that the elderly vaccination programme is cost-effective Preschool are children aged 2–4 years, primary are children aged 5–10 years, and secondary are adolescents aged 11–16 years. Paediatric vaccination coverage is set at 70% and is reached at speeds consistent with those observed during the full rollout of preschool and the pilot rollout of school-age children. Three different incremental cost-effectiveness ratio thresholds are considered: £30 000 per QALY (red; cost-effective), £20 000 per QALY (blue; very cost-effective under current protocol), and £15 000 per QALY (green; cost-effective under proposed protocol). A relative direct effectiveness of 1 corresponds to the LAIV having the same whole-season direct effectiveness as the IIV (base case value), and a relative direct effectiveness of 0 corresponds to the LAIV having no effect on influenza epidemiology and is therefore equivalent to the cost-effectiveness of the elderly programme with no paediatric coverage. LAIV=live attenuated influenza vaccine. QALY=quality-adjusted life-year. IIV=inactivated influenza vaccine.