Reply to letter from Alvarez et al.

KEYWORDS:

• Influenza
• vaccination
• Spain
• Quadrivalent influenza vaccine (QIV)
• costeffectiveness
• cell-based vaccines
• high-risk
• burden of illness

Dear Dr. Ellis,

Thank you for the opportunity to address the concerns outlined in the letter from Alvarez et al. in response to our study “Estimated Cost-effectiveness and Burden of Disease Associated with Quadrivalent Cell-based and Egg-based Influenza Vaccines in Spain” that concludes cell-based quadrivalent influenza vaccine (QIVc) offers a cost-effective alternative to egg-based quadrivalent influenza vaccine (QIVe) and should be considered as an alternative vaccine to QIVe for people aged 9–64 at high risk of influenza complications in Spain.¹

Alvarez et al. correctly point out that the results are highly sensitive to the benefit of QIVc over QIVe and that vaccine effectiveness (VE) should ideally be based on systematic review or meta-analysis and subject to sensitivity analyses.

When our economic evaluation was conducted, the study by Boikos et al.² (now published in a peer-review journal³) was the only study of its kind. It is based on influenza-like-illness (ILI) defined through diagnosis codes and included over 60,000 patients. Alvarez et al. suggest alternative sources for meta-analyses, but these are unsuitable because:

• Izurieta et al. ^4,5 only considered a population aged 65 yrs+ whereas our study was in 9–64 y olds at high-risk of influenza complications.

• The population in DeMarcus et al.⁶ was Department of Defense (DoD) beneficiaries who presented to a military treatment facility. These were DoD dependants and therefore unrepresentative of the high-risk population of interest.

• Klein et al.⁷ included patients positive for influenza by polymerase chain reaction (PCR) rather than ILI and aged 4–64 y.

• Bruxvoort et al.⁸ used hospitalization for laboratory-confirmed influenza rather than ILI and included only 75 QIVc subjects under 65 y old (aged 4 to 64).

Combining these datasets within a single analysis would have introduced unacceptable levels of heterogeneity and made the data no longer pertinent to the decision problem, as different severity measures, age groups and risk groups are well known to have different outcomes.⁹

Alvarez et al. incorrectly suggest we did not undertake sensitivity analyses regarding VE, as both one-way deterministic sensitivity analysis and probabilistic sensitivity analyses were conducted for VE. In response to the concerns raised by Alvarez et al. we have conducted additional analyses which shows:

• At the Spanish willingness-to-pay threshold of €22,000 per QALY,¹⁰ QIVc would be cost-effective compared to QIVe down to a relative VE (rVE) of 17.5% across all age-groups.

• If we use the value of 18.8% in the 9–17 age-group and 26.8% in the 18–64 age-group, the ICER is €13,551 per QALY from a medical cost perspective and dominant from a societal perspective.

• If we consider only the 9–17 age-group and use the value of 18.8% in this age group, the ICER is €17,898 from a medical cost perspective and dominant from a societal perspective.

We, therefore, refute the claims that our study does not meet the minimal standards for cost-effectiveness analysis. It is based on the best data available to address the decision problem, is clear and transparent, and includes extensive sensitivity and scenario analyses. These demonstrate the conclusions are both valid and robust.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.