Indirect Influenza Vaccine Effectiveness Under Randomized Conditions: A Systematic Review and Meta-Analysis

Indirect vaccine effectiveness (IVE) estimates are vulnerable to bias, and high-quality evidence on influenza IVE remains limited. We conducted a systematic review and meta-analysis of randomized controlled trials (RCTs) estimating influenza IVE.

We searched MEDLINE and EMBASE through 9 December 2025, for RCTs conducted in non-institutional settings estimating IVE by comparing reverse transcription-polymerase chain reaction-confirmed influenza illness among unvaccinated individuals in vaccinated versus unvaccinated households or communities. Searches were restricted to English-language publications.

Eight articles describing twelve RCTs were included, comprising 10 cluster-randomized trials designed to estimate IVE and 2 individual-randomized trials assessing IVE among household contacts. Risk of bias was low. Across trials conducted between 2008 and 2013, estimates of direct vaccine effectiveness (DVE) were moderate, whereas IVE estimates were lower and often centered near no effect, with the CIs for 9 of 12 trials including zero. The pooled DVE was 44.1% (95% CI: 29.6%, 55.6%), and the pooled IVE was 13.7% (95% CI: 1.5%, 24.5%). Sensitivity analyses excluding pandemic outcomes yielded minimally higher point estimates and did not materially change the findings. Between-study heterogeneity was moderate. There was no evidence of publication bias. Certainty of evidence was rated as moderate.

This review provides a methodologically rigorous benchmark of IVE measurable under randomized conditions. While pooled DVE was moderate, pooled IVE was lower, and most individual trials did not demonstrate statistically significant indirect effects. The findings indicate that individual vaccination is the most reliable means of preventing influenza, rather than depending on indirect or community-level protection.