The Clinical Impact and Cost Effectiveness of Quadrivalent Versus Trivalent Influenza Vaccination in Finland

Abstract

Background: Trivalent influenza vaccines encompass one influenza B lineage; however, predictions have been unreliable on which of two antigenically distinct circulating lineages will dominate. Quadrivalent seasonal influenza vaccines contain strains from both lineages. This analysis assesses the cost effectiveness of switching from trivalent inactivated influenza vaccination (TIV) in Finland to quadrivalent vaccination, using inactivated (QIV) or live-attenuated (Q-LAIV) vaccines.

Methods: A transmission model simulated the dynamics of influenza infection while accounting for indirect (herd) protection. Prior distributions for key transmission parameters were repeatedly sampled and simulations that fitted the available information on influenza in Finland were recorded. The resulting posterior parameter distributions were used in a probabilistic sensitivity analysis in which economic parameters were sampled, simultaneously encompassing uncertainty in the transmission and economic parameters. The cost effectiveness of a range of trivalent and quadrivalent vaccine policies over a 20-year time horizon was assessed from both a societal and payer perspective in 2014 Euros.

Results: The simulated temporal incidence pattern of symptomatic infections corresponded well with case surveillance data. A switch from the current TIV to Q-LAIV in children (2 to <18 years) and to QIV in other ages was estimated to annually avert approximately 76,100 symptomatic infections (95 % range 36,700-146,700), 11,500 primary care consultations (6100-20,000), 540 hospitalisations (240-1180), and 72 deaths (32-160), and was cost-saving relative to TIV (€374 million averted [€161-€752], in 2014 Euros, discounted at 3 %). This scenario had the highest probability of being the most cost-effective scenario considered.

Conclusions: This analysis demonstrates that quadrivalent vaccination is expected to be highly cost effective, reducing the burden of influenza-related disease.

Fig. 1

Vaccine uptake, from a TNS Gallup Oy survey, sampled in the probabilistic sensitivity analysis

Fig. 2

Monthly incidence of symptomatic influenza, averaged over all runs and normalised to the 2000/2001 to 2007/2008 overall average. Comparison of the normalised modelled monthly incidence of symptomatic influenza infection to normalised National Institute for Health and Welfare (THL) surveillance reports [8] of symptomatic influenza. Figure derived from THL data, with permission

Fig. 3

Distribution of parameter samples over 8998 runs: calibrated distributions, by virus, of selected parameters contributing to the basic reproduction number (R ₀) and resultant R ₀ distributions

Fig. 4

Distribution of basic reproduction number (R ₀) parameter samples over 8998 runs: comparison between initial input distributions and updated (calibrated) distributions, by virus

Fig. 5

Discounted incremental costs (2014 Euros) and QALYS of five vaccination scenarios compared with the no vaccination scenario. Bars represent the range in which 95 % of simulations fell. Scenario names in the key are defined in Table 3. QALYs quality-adjusted life-years, TIV trivalent inactivated vaccine, QIV quadrivalent inactivated vaccine, Q-LAIV quadrivalent live-attenuated influenza vaccine

Fig. 6

CEACs and CEAFs for the scenarios presented in Sect. 2.3.1, both including all calculated costs and with productivity costs excluded. Scenario names in the key are defined in Table 3. ENB expected net benefit. CEAC cost-effectiveness acceptability curve, CEAF cost-effectiveness acceptability frontier, QALY quality-adjusted life-year, TIV trivalent inactivated vaccine, QIV quadrivalent inactivated vaccine, Q-LAIV quadrivalent live-attenuated influenza vaccine