Abstract
This paper studies the direct and indirect effectiveness of Covid-19 vaccines among vaccinated healthcare workers and their unvaccinated adult household members in a mass vaccine program in Finland.
Methods We used national databases that record all polymerase chain reaction (PCR)-confirmed SARS-CoV-2 infections and mRNA-based (BNT162b2 by Pfizer-BioNTech or mRNA-1273 by Moderna) vaccine doses administered in Finland since the beginning of the epidemic. These data were merged with administrative full population datasets that include information on each person’s occupation and unique identifiers for spouses living in the same household. To estimate the direct and indirect effectiveness of mRNA-based vaccines in a household setting, we compared the cumulative incidence of PCR-confirmed SARS-CoV-2 infections between vaccinated and unvaccinated healthcare workers as well as between their unvaccinated spouses.
Findings Our estimates imply indirect effectiveness of 8.7% (95% CI: -28.9 to 35.4) two weeks and 42.9% (95% CI: 22.3 to 58.1) 10 weeks after the first dose. The effectiveness estimates for unvaccinated household members are substantial, but smaller than the direct effect and occur more gradually among unvaccinated household members than among vaccinated individuals.
Interpretation Our results suggest that mRNA-based vaccines do not only prevent SARS-CoV-2 infections among vaccinated individuals but lead to a substantial reduction in infections among unvaccinated household members. The results are consistent with the notion that mRNA-based vaccines affect susceptibility in vaccinated individuals and prevent transmission from vaccinated to unvaccinated individuals.
Governments around the world are hoping to lift Covid-19 restrictions as vaccination coverage increases. Simultaneously, many under-vaccinated countries and territories have yet to design and initiate their mass vaccination campaigns. Vaccines have been shown to be effective in reducing SARS-CoV-2 infections among vaccinated individuals both in clinical trials and in nationwide mass vaccination settings.1–4 However, there is only little real-world descriptive evidence about the indirect effects of Covid-19 vaccinations.5–8
We studied the direct and indirect effectiveness of Covid-19 vaccines in reducing the cumulative risk of infection among vaccinated healthcare workers and their unvaccinated adult household members in a mass vaccine program in Finland. We used national databases that record all polymerase chain reaction (PCR)-confirmed SARS-CoV-2 infections and mRNA-based (BNT162b2 by Pfizer-BioNTech or mRNA-1273 by Moderna) vaccine doses administered in Finland since the beginning of the epidemic. These data were merged with administrative full population datasets that include information on each person’s occupation and unique identifiers for spouses living in the same household. To estimate the direct and indirect effectiveness of mRNA-based vaccines in a household setting, we compared the cumulative incidence of PCR-confirmed SARS-CoV-2 infections between vaccinated and unvaccinated healthcare workers as well as between their unvaccinated spouses (see Supplementary Appendix Table A2 for the covariate balance between the vaccinated and unvaccinated groups). The estimates were adjusted for the state of the epidemic (calendar time fixed effects), age and sex.
Figure 1 shows the effectiveness estimates of mRNA-based vaccines for preventing Covid-19 infections among working-age healthcare workers and their spouses between the start of the mass vaccination program (December 27, 2020) and March 24, 2021 in Finland. We report estimates by follow-up week after receiving the first dose. We included only persons who had not been infected before the event date. For unvaccinated individuals, we randomly assigned a placebo event date for vaccine administration during the observation period (see the Supplementary Appendix, Section D, for details). Using this novel approach, we ensure that the vaccine effectiveness estimates are not biased by the changing nature of the epidemic between the treatment and control groups during the follow-up period. Our results remain qualitatively unaltered if we follow all individuals in the control group from the beginning of the mass vaccination program (Supplementary Appendix, Figure A3). We used a log-binomial model as it allows direct estimation of vaccine effectiveness in terms of risk reduction.
Our estimates imply a gradual increase in direct effectiveness over time (Figure 1, Panel A): 26.8% (95% CI: 7.5 to 42.1) two weeks and 69.0% (95% CI: 59.2 to 76.3) 10 weeks after the first dose. These estimates are consistent with the results from previous clinical trials and observational studies. Our results document vaccine effectiveness in a real-world setting, in which it is difficult to separate the longer-term effects of first- and second-dose vaccinations. Over 40% of individuals included in our sample received their second dose four weeks after the first dose (Supplementary Appendix, Figure A1).
Our estimates imply indirect effectiveness of 8.7% (95% CI: -28.9 to 35.4) two weeks and 42.9% (95% CI: 22.3 to 58.1) 10 weeks after the first dose (Figure 1, Panel B). The effectiveness estimates for unvaccinated household members are substantial, but smaller than the direct effect and occur more gradually among unvaccinated household members than among vaccinated individuals.
Our results suggest that mRNA-based vaccines do not only prevent SARS-CoV-2 infections among vaccinated individuals but lead to a substantial reduction in infections among unvaccinated household members. The results are consistent with the notion that mRNA-based vaccines affect susceptibility in vaccinated individuals and prevent transmission from vaccinated to unvaccinated individuals.9
The present study is uniquely suited to evaluating the effect of mRNA-based Covid-19 vaccines on secondary transmission from vaccinated to unvaccinated household members. We were able to merge individual-level health records to full population data containing detailed information about individuals’ occupations and household IDs, allowing us to estimate the indirect effects of vaccines in a large sample of couples with discordant vaccination status. Our study is limited by the fact that SARS-CoV-2 virus exposure may differ between vaccinated and unvaccinated healthcare workers as persons directly involved with Covid-19 patients are more likely to be vaccinated. Taken together, our result may provide support for designing effective vaccination strategies, societal control measures and travel recommendations for vaccinated individuals.
Data Availability
This paper uses administrative health care and employment data maintained by the Finnish Institute for Health and Welfare and Statistics Finland. Health care data is regulated under the Act on the Secondary Use of Health and Social Data (552/2019) and can be obtained by sending a direct request to the Finnish Institute for Health and Welfare (https://thl.fi/en). The Finnish Longitudinal Employer-Employee Data (FOLK) can be obtained by sending a direct request to Statistics Finland (https://www.stat.fi). The authors are willing to assist in making data access requests.
Footnotes
E-mail: jussipekka.salo{at}helsinki.fi
E-mail: milla.hagg{at}helsinki.fi
E-mail: mika.kortelainen{at}utu.fi
E- mail: tuija.leino{at}thl.fi
E-mail: tanja.saxell{at}gmail.com
E-mail: markku.siikanen{at}vatt.fi
E-mail: lauri.saaksvuori{at}thl.fi
This research was supported by the InFLAMES and INVEST Flagship Programmes of the Academy of Finland. Dr. Kortelainen declares grant to his employer, but no personal support or financial relationship, from Pfizer during the conduct of the study. Other authors have no conflict of interest to declare.