The Number Needed to Vaccinate (NNV) against herpes zoster: a systematic review with meta-analysis

Abstract

Objectives

The number needed to vaccinate (NNV) quantifies the effectiveness of vaccination programs. We summarised the published data on NNV against herpes zoster to inform vaccination policies.

Methods

We systematically identified studies based on a priori established and registered methods. The main outcomes were the NNV against herpes zoster infection, hospitalisation and mortality. Where appropriate, we conducted meta-analyses using inverse variance, random-effects models, pooling estimated NNV with associated 95% confidence interval (CI). Statistical heterogeneity between pooled estimates was calculated using the I² statistic.

Results

Out of 229 unique citations, we included eight nonrandomized studies. Among 50+ year-olds, the NNV against herpes zoster infection using the recombinant subunit vaccine was 11 (95%CI 8–14; I² = 0%; 3 studies) and variable (I² = 94.4%; 7 studies) using live attenuated vaccine, ranging from 10 (95%CI 1–19) to 58 (95%CI 49–67). Among 65+ year-olds, the NNV against herpes zoster infection using the recombinant subunit vaccine was 12 (95%CI: 9–15; I² = 0%; 2 studies) and variable (I² = 98.5%; 4 studies) using live attenuated vaccine, ranging from 14 (95%CI 5–23) to 75 (95%CI 66–84). The NNV against herpes zoster hospitalisation among 65+ year-olds using the live attenuated vaccine was 280 (95%CI 209–352; I² = 0%; 2 studies). There was a paucity of data to inform other meta-analyses.

Conclusion

Evidence on the NNV against herpes zoster is scarce. Vaccination with the recombinant subunit herpes zoster vaccine may be more effective than with the live attenuated vaccine in preventing infection among 50+ year-olds. More studies are needed for a stronger evidence base for decision-making.

Keywords:

• Number needed to vaccinate
• Herpes zoster
• systematic review
• meta-analysis

Acknowledgments

We thank Hal Loewen, MLIS (Neil John Maclean Health Sciences Library, University of Manitoba, Winnipeg, Manitoba, Canada) for peer reviewing the MEDLINE search strategy.

Author contributions

Design (GNO); Data collection (GNO, YA, VKR, ML, CJN); Data analysis/interpretation (GNO); Draft manuscript (GNO); Manuscript revisions (GNO, YA, VKR, ML, CJN, AMAS); Approval for submission (GNO, YA, VKR, ML, CJN, AMAS).

Disclosure statement

No potential conflict of interest was reported by the authors.