A Meta-Analysis of Influenza Vaccination Following Correspondence: Considerations for COVID-19

Abstract

AbstractBackgroundHigh vaccination rates are needed to protect against influenza and to end the COVID-19 pandemic. Health authorities need to know if supplementing mass communications with direct correspondence to the community would increase uptake.ObjectivesThe primary objective is to determine if sending a single written message directly to individuals increases influenza vaccine uptake, and a secondary objective is to identify any identified content shown to increase influenza vaccine uptake.MethodsPubMed, PsycInfo and Web of Science were searched for English language RCTs testing a single correspondence for members of the community in OECD countries to obtain influenza vaccination. A meta-analysis with inverse-variance, random-effects modelling was used to estimate a mean, weighted risk ratio effect size measure of vaccine uptake. Studies were quality assessed and analysis was undertaken to account for potential publication bias.ResultsTwenty-two randomized controlled trials were included covering 37 interventions. Of the 37 interventions, 32 (86%) report an increase in influenza vaccination rates. A formal meta-analysis shows that sending a single written message increases influenza vaccine uptake by 18% (RR = 1.18, 95%CI [1.13-1.22], Z = 8.56, p < .001) relative to the no contact comparator group. Analysis shows that the intervention is effective across correspondence type, age group, time, and location, and after allowing for risk of publication bias.LimitationsThe review was restricted to English language publications, and the generalizability of results across the OECD may be questioned.Conclusions and implicationsThe implication for public health authorities organizing vaccination programs for influenza, and arguably also for COVID-19, is that sending written vaccination correspondence to members of the community is likely to increase uptake.The review was not registered nor was a protocol prepared due to time sensitivity.