How does the number needed to vaccinate (NNV) for Shingrix change by age group and immunocompromise status?
Executive summary
The number needed to vaccinate (NNV) for Shingrix cannot be pinned to a single figure from available public reports because NNV depends on vaccine efficacy (VE) and the baseline risk of herpes zoster in the population — parameters that vary by age and by type/degree of immunocompromise (and the sources provide VE estimates but not the consistent incidence rates needed to compute NNV precisely) [1] [2]. What the public record does allow is a clear qualitative map: VE is very high in immunocompetent older adults and more variable in immunocompromised groups, while baseline risk of zoster rises with age and can be as high or higher in some younger immunocompromised patients, so NNV will fall (become more favorable) when baseline risk is higher and VE remains substantial, and will rise when VE is reduced or baseline risk is lower [3] [1] [2].
1. What “NNV” means and why age and immune status matter
NNV is 1 divided by the absolute risk reduction (ARR) achieved by vaccination, so it depends on two inputs: the vaccine’s relative efficacy and the baseline incidence of disease in the unvaccinated group; higher baseline incidence or higher VE produce a smaller (better) NNV, while lower baseline incidence or lower VE produce a larger (worse) NNV (this mathematical relationship underlies all comparative claims about NNV) [1]. The CDC and ACIP materials stress that shingles risk increases with age and that adults ≥50 are the primary immunocompetent target because incidence and complications climb with advancing age, a key driver making NNV lower in older age bands if VE is high [3] [1].
2. What the evidence says about vaccine efficacy by age in immunocompetent adults
Prelicensure trials reported very high VE in immunocompetent adults: about 97% against zoster for ages 50–59 and 60–69, and roughly 91% for those 70 and older, with some waning by year four but sustained substantial protection — these high relative reductions imply favorable NNVs in older immunocompetent adults, provided baseline zoster incidence in those age bands is not trivial [4]. Because the sources do not enumerate age-specific incidence numbers in the same tables they provide VE, an exact NNV for “age 50–59” versus “70+” cannot be derived from the cited documents alone; the required ARR calculation is dependent on baseline incidence figures not supplied in these excerpts [4] [3].
3. How immunocompromise alters VE and therefore shifts NNV
ACIP and trial summaries demonstrate that VE among immunocompromised groups is heterogeneous: example VE estimates include 68.2% for autologous hematopoietic cell transplant recipients, and substantially higher point estimates (87.2% and 90.5%) in post hoc analyses for hematologic malignancy patients and those with “potential immune-mediated diseases,” respectively — the lower VE in some groups would increase NNV compared with a group with higher VE, all else equal [1] [2]. Importantly, CDC and ACIP note that some younger adults with specific immunocompromising conditions have zoster risk comparable to or greater than that of the general population aged >50, which means that even with somewhat lower VE, a high baseline risk in an immunocompromised younger adult could produce an NNV similar to or better than that of older immunocompetent adults [2] [1].
4. Putting the pieces together: qualitative NNV comparisons across groups
Synthesis of the sources supports three practical patterns: immunocompetent adults 50–69 with very high VE will generally have low NNV because high efficacy combines with appreciable baseline incidence in older age [4] [3]; adults ≥70 have slightly lower VE so NNV may be a bit higher than for 50–69 but is tempered by substantially higher baseline risk of disease and complications, which can preserve a favorable NNV [4] [3]; immunocompromised adults show variable VE — where VE is high and baseline risk is high (e.g., some hematologic malignancy groups) NNV will be low, but in groups with lower VE (e.g., autologous HCT with VE ≈68%) NNV will be larger unless baseline incidence is exceptionally high, a determination that requires condition-specific incidence data not given in the provided documents [1] [2].
5. Limitations, uncertainties, and the influence of messaging
The available public documents provide VE point estimates for age bands and for certain immunocompromised cohorts, but they do not give uniform, matched baseline incidence rates across those same cohorts needed to compute exact NNVs, so any numeric NNV presented without those incidence inputs would be speculative based on the sources provided [1] [2] [4]. Stakeholders have clear incentives: regulators and CDC emphasize population benefit and safety in expanding recommendations [1], while manufacturer communications highlight the size of the newly eligible population and the vaccine’s role in addressing unmet need — readers should recognize those differing agendas when interpreting claims about how “cost-effective” or impactful vaccination will be for particular subgroups [5].