Fig. 3. Vaccine effectiveness among fully (double) vaccinated healthcare workers and their unvaccinated partners and children living in the same household after the second vaccine dose.

This figure plots the vaccine effectiveness estimates by week after the second dose of mRNA-based vaccination. The dependent variable is a polymerase chain reaction (PCR)-confirmed SARS-CoV-2 infection as recorded in the Finnish National Infectious Diseases Register. a Direct effectiveness estimates (relative risk reduction) for vaccinated individuals compared to an unvaccinated control group, which is constructed by randomly assigning the beginning of a follow-up period for unvaccinated individuals. N = 216,557 healthcare worker observations. b Indirect effectiveness estimates (relative risk reduction) for the unvaccinated partners (including cohabiting partners) of fully vaccinated individuals who lived in the same household as the vaccinated person as of December 31, 2019. N = 110,426 partner observations. c Indirect effectiveness estimates (relative risk reduction) for unvaccinated household members aged from 3 to 18 years who lived in the same household as the vaccinated person as of December 31, 2019. N = 144,935 child and adolescent observations. d Effectiveness estimates (relative risk reduction) separately for unvaccinated children aged from 3 to 12 years and adolescents aged from 13 to 18 years who lived in the same household as the vaccinated person as of December 31, 2019. N = 91,031 child (3–12 years) observations, 53,904 adolescent (13–18 years) observations. c, d Stack the estimates from follow-up weeks 2–5 into a single estimate as there are too few PCR-confirmed SARS-CoV-2 infections to make any inference about the short-term effectiveness of mRNA-based vaccination in this age group. The estimate for weeks 2–5 represents the average vaccine effectiveness in weeks 2, 3, 4, and 5. The coefficients in all panels are estimated using a log-binomial regression model and individual-date data collapsed to individual-week data. See section Methods, Estimation samples, for details of the sample restrictions and estimation method. The relative risk reduction is presented as a percentage in data points ±95% confidence intervals. The solid black dots show relative risk reduction by week after receiving the second dose of mRNA-based vaccination. The error bars denote the 95% confidence intervals clustered at the individual level using endpoint transformation of the originally estimated confidence intervals.