Comment on “Cost-Effectiveness Analysis of Herpes Zoster Vaccination in 50- to 85-Year-Old Immunocompetent Belgian Cohorts: A Comparison between No Vaccination, the Adjuvanted Subunit Vaccine, and Live-Attenuated Vaccine”

Dear Editor,

In a recent publication, Pieters et al. [1] investigated the cost effectiveness of the recombinant zoster vaccine (RZV) and of zoster vaccine live (ZVL) for the prevention of herpes zoster (HZ) in immunocompetent older adults in Belgium. While valuable in being the first publication examining the cost effectiveness of RZV using local Belgian data, the publication underestimates the public health benefits of RZV vaccination as well as its cost-effectiveness levels for Belgium, primarily because (a) it applies HZ incidence rates adjusted to immunocompetent individuals using a simplistic and biased calculation and (b) it applies the aforementioned adjustments to RZV versus no vaccination comparisons although RZV is indicated for both immunocompetent and immunocompromised individuals. In the remainder of our communication, we briefly elaborate on these two points.

On point (a), we observe that Pieters et al. [1] attempt to adjust the overall HZ incidence rates previously reported in Bilcke et al. [2], originally published in KCE (Belgian Health Care Knowledge Centre) Report 151 [3], which include immunocompetent and immunocompromised subjects, to immunocompetent subjects only, by simplistically multiplying the overall HZ general practitioner visit and hospitalisation rates by the proportions of ambulatory and hospitalised HZ cases in immunocompetent individuals, respectively. This approach underestimates the incidence rates in the immunocompetent population. Note the overall incidence is a combination of the incidence in the immunocompetent and immunocompromised populations as follows:

$$\text{IR}(\text{ALL})=\text{P}(\text{IC}\_\text{FREE})\times \text{IR}(\text{IC}\_\text{FREE})+[1-\text{P}(\text{IC}\_\text{FREE})]\times \text{RR}\times \text{IR}(\text{IC}\_\text{FREE})$$

where IR(ALL) denotes the overall HZ incidence rate, IR(IC_FREE) is the HZ incidence rate for immunocompetent (IC-free) subjects, P(IC_FREE) is the proportion of the IC-free segment in the general population, and RR is the average HZ risk ratio aggregated over all immunocompromising conditions.

Re-arranging the formula above, we observe that

$$\text{IR}(\text{IC}\_\text{FREE})=\text{IR}(\text{ALL})/(\text{P}(\text{IC}\_\text{FREE})+[1-\text{P}(\text{IC}\_\text{FREE})]\times \text{RR})$$

Proper adjustments would therefore rely on knowledge of the true proportion of immunocompetent individuals in the population, as well as on appropriate risk ratios, defined as the risk of HZ in the immunocompromised population over the risk of HZ in the immunocompetent population.

On point (b), we note that the approach of evaluating the cost effectiveness of RZV in immunocompetent individuals only underestimates the overall value of RZV, as RZV is also indicated for adults immunocompromised due to illness or immunosuppressed due to treatment at the time of vaccination [4], whereas ZVL is generally contra-indicated in subjects with immunodeficiency or undergoing immunosuppressive therapy [5].

In conclusion, adjusting the overall incidence rates of HZ to account for immunocompetent subjects is unnecessary for RZV as the vaccine is indicated for the prevention of HZ in adults 18 years of age or older who are immunodeficient or immunosuppressed due to disease or therapy. In the unlikely event that a dedicated analysis on the immunocompetent segment of the population were mandated, proper methodology and sensible adjustment of the overall incidence rates of HZ should be employed.

Declarations

Funding

GlaxoSmithKline Biologicals SA funded all costs associated with the development and publication of this letter.

Conflict of interest

NG and NL are employees of GSK and hold shares in GSK. MLN is an AIXIAL Group consultant on behalf of GSK. AIXIAL received fees from GSK for this study. The authors declare no other financial and non-financial relationships and activities and no other conflicts of interest.

Author contributions

All authors participated in the development of this publication and gave final approval before submission.

Ethics approval

Not applicable.

Informed consent

Not applicable.

Data availability

Not applicable.