Vaccination strategies for future influenza pandemics: a severity-based cost effectiveness analysis

Abstract

Background: A critical issue in planning pandemic influenza mitigation strategies is the delay between the arrival of the pandemic in a community and the availability of an effective vaccine. The likely scenario, born out in the 2009 pandemic, is that a newly emerged influenza pandemic will have spread to most parts of the world before a vaccine matched to the pandemic strain is produced. For a severe pandemic, additional rapidly activated intervention measures will be required if high mortality rates are to be avoided.

Methods: A simulation modelling study was conducted to examine the effectiveness and cost effectiveness of plausible combinations of social distancing, antiviral and vaccination interventions, assuming a delay of 6-months between arrival of an influenza pandemic and first availability of a vaccine. Three different pandemic scenarios were examined; mild, moderate and extreme, based on estimates of transmissibility and pathogenicity of the 2009, 1957 and 1918 influenza pandemics respectively. A range of different durations of social distancing were examined, and the sensitivity of the results to variation in the vaccination delay, ranging from 2 to 6 months, was analysed.

Results: Vaccination-only strategies were not cost effective for any pandemic scenario, saving few lives and incurring substantial vaccination costs. Vaccination coupled with long duration social distancing, antiviral treatment and antiviral prophylaxis was cost effective for moderate pandemics and extreme pandemics, where it saved lives while simultaneously reducing the total pandemic cost. Combined social distancing and antiviral interventions without vaccination were significantly less effective, since without vaccination a resurgence in case numbers occurred as soon as social distancing interventions were relaxed. When social distancing interventions were continued until at least the start of the vaccination campaign, attack rates and total costs were significantly lower, and increased rates of vaccination further improved effectiveness and cost effectiveness.

Conclusions: The effectiveness and cost effectiveness consequences of the time-critical interplay of pandemic dynamics, vaccine availability and intervention timing has been quantified. For moderate and extreme pandemics, vaccination combined with rapidly activated antiviral and social distancing interventions of sufficient duration is cost effective from the perspective of life years saved.

Figure 1

Synopsis of pandemic scenarios and intervention strategies. A synopsis of three pandemic scenarios and plausible intervention strategies. Social distancing (SD) intervention indicates school closure (SC) and community contact reduction (CCR). Rigorous social distancing indicates the addition of workforce reduction (WR) to the SC and CCR interventions.

Figure 2

Cost effectiveness plane for intervention strategies. Each intervention strategy is plotted with horizontal position according to relative cost in dollars per member of population compared to no intervention, and with vertical position according to the number of life-years saved (LYS) per 10,000 population. The lower figure contains an enlarged view around the axis to clarify the position of interventions for moderate and mild pandemics. Colours denote pandemic severity: red for extreme (R₀ = 2.7, CFR = 1.5%), green for moderate (R₀ = 1.9, CFR = 0.25%) and blue for mild (R₀ = 1.5, CFR = 0.03%). Crosses indicate interventions without vaccination, diamonds indicate vaccination interventions. Interventions are labelled as for Figure 1, abbreviations used in the figure legend are SC for school closure, CCR for community contact reduction, AV for antiviral treatment and household prophylaxis, and WR for workforce reduction.

Figure 3

Epidemic curves with and without vaccination for moderate pandemics. Daily incidence curves are shown for six intervention strategies for a mild pandemic (R₀ = 1.9) Figure legend text. The left panel shows three strategies that do not include vaccination while the right panel shows three strategies than include vaccination. The composition of each strategy is given in the figure legend along with the final attack rate for that strategy. Points along the horizontal (time) axis where social distancing interventions are relaxed or vaccination begins are marked with a vertical line.

Figure 4

Cost component breakdown for intervention strategies. The total cost for each intervention strategy and pandemic severity scenario is broken down into 5 component costs, with the width of each coloured bar indicating the percentage of the total cost constituted by each component – health care costs (dark blue), including GP visits and hospitalisation; antiviral costs (red) including pharmaceutical, dispensing and stockpile renewal costs; vaccination costs (green) including vaccine development, production and distribution; lost productivity due to social distancing and illness (purple); and productivity loss due to death (light blue). costs and the percentage of each component is indicated by a coloured bar. Note that all total costs are scale to have the same width, so absolute widths are not comparable between strategies.