National epidemiological analysis of the association of COVID-19 vaccination and incidence of COVID-19 cases in Canada, January to August 2021.

Abstract

In December 2020, Canada began its coronavirus disease 2019 (COVID-19) vaccine rollout campaign. Canadians were vaccinated with differing time intervals between doses, vaccine products and vaccine schedules, based on age, timing of vaccination and jurisdiction. The objective of this study is to describe the epidemiology and association between the incidence of COVID-19 cases following vaccination, time since completion of primary series, time between doses and/or product combination and probability of developing severe outcomes. The national COVID-19 case data and vaccination coverage data were extracted from the National COVID-19 Surveillance System, and the Canadian COVID-19 Vaccination Coverage Surveillance System. Population estimates from Statistics Canada were used as denominators for rates and for number of people "not fully vaccinated". Two binomial generalized linear models were constructed for analysis. Within the analysis period, fully vaccinated (i.e. completed primary series) cases (n=17,206) were more commonly female and older, and had fewer reported severe outcomes relative to not fully vaccinated cases (n=615,999). Episode date of fully vaccinated cases most frequently occurred two months after receiving their second dose, and time-between doses of 29-49 and 50-77 days were most common. The probability of becoming a detected COVID-19 case in not fully vaccinated individuals was higher than those fully vaccinated. Those receiving two doses of AstraZeneca and those with shortest time intervals between doses had higher probabilities of becoming COVID-19 cases. Findings from Canada's national surveillance systems support that being fully vaccinated against COVID-19, having a longer time interval between doses and receiving a messenger ribonucleic acid (mRNA) COVID-19 vaccine schedule compared to other vaccines reduce the probability of becoming a case, using data from January to August 2021.