COVID-19 outcome trends by vaccination status in Canada

Abstract

Background: The coronavirus disease 2019 (COVID-19) pandemic in Canada has evolved rapidly. Since late 2020, COVID-19 vaccines have been relied on to protect against severe outcomes in the presence of circulating variants of concern (VOC).

Objective: This surveillance report provides a retrospective descriptive analysis of national trends in COVID-19 cases and severe outcomes by vaccination status, contextualizing trends against case demographics and circulating VOCs, from December 2020 to January 2022.

Methods: Case and vaccination coverage surveillance data were obtained from the National COVID-19 Case Dataset and the Canadian COVID-19 Vaccination Coverage Surveillance System for 12 of 13 provinces and territories. Descriptive analyses were produced to describe trends over time among individuals aged 12 years and older by COVID-19 outcome, vaccination status, and demographics. Age-standardized and age-stratified incidence rates and incidence rate ratios were computed for cases, hospitalizations, and deaths.

Results: From mid to late-2021, incidence rates for cases and severe outcomes were consistently lowest among those with a completed primary series and highest among those who were unvaccinated. Unvaccinated individuals were much more likely to be hospitalized or to die compared to those with a completed primary series in all variant periods. Age-specific rates of severe outcomes were consistently highest among those aged 80 years and older across all vaccination statuses.

Conclusion: Vaccination remains one of the most important public health interventions, particularly among older adults, to protect against COVID-19 severe outcomes as the pandemic evolves. Routine monitoring of COVID-19 outcomes by vaccination status can identify changes in COVID-19 epidemiology and inform public health action and policy.

Introduction

The coronavirus disease 2019 (COVID-19) pandemic has been one of the most significant public health crises in the last century, resulting in increased morbidity, mortality, and social and economic disruption in Canada and worldwide ^{Footnote 1Footnote 2}. Until vaccines were first authorized in Canada on December 9, 2020, broad and stringent public health measures (PHMs) were heavily relied on to slow transmission of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and mitigate its impacts on health and society ^{Footnote 1Footnote 3}.

From December 2020 to January 2022, the COVID-19 pandemic in Canada evolved rapidly. Variants of the SARS-CoV-2 wild-type virus, including Alpha, Beta, Gamma, Delta, and Omicron, emerged, with differential impacts on COVID-19 outcomes ^{Footnote 3}. In Canada, public health falls under provincial/territorial jurisdiction and PHMs (e.g., travel restrictions, work/school closures, and personal protective measures, such as masking, and physical distancing) ^{Footnote 3}, testing strategies, and vaccine policies varied across provinces and territories (hereafter referred to as jurisdictions). Confirmatory COVID-19 testing was made broadly available across jurisdictions ^{Footnote 4}. Canadian COVID-19 vaccination campaigns began on December 14, 2020, initially prioritizing vulnerable and at-risk populations ^{Footnote 3Footnote 5}. Between March and November 2021, jurisdictions expanded vaccination eligibility from older adults to children older than five years, following changes to the availability of vaccines and recommendations from the National Advisory Committee on Immunization (NACI) ^{Footnote 5Footnote 6}. From September to December 2021, NACI recommended a booster dose of an authorized mRNA COVID-19 vaccine for key populations to address waning immunity and suboptimal primary series vaccine effectiveness ^{Footnote 7Footnote 8Footnote 9}. By January 1, 2022, 88% of people aged 12 years or older had received a primary COVID-19 vaccine series, and 19% had received a primary series with one additional dose ^{Footnote 10}.

This report provides a retrospective, descriptive analysis of national trends in COVID-19 outcomes by vaccination status among individuals aged 12 years and older, contextualized by variant and case demographic characteristics, from December 2020 to January 2022.

Methods

Data sources

Data were obtained from the National COVID-19 Case Dataset, a case-based surveillance system that collects data on demographics, clinical status and outcomes, risk factors, vaccination, and variant lineages of COVID-19 cases in Canada. Jurisdictions report case data electronically to the Public Health Agency of Canada (PHAC) at varying frequencies. Data are subsequently mapped and stored in a Postgres (PostgreSQL) database maintained by PHAC (Metabase).

Data on vaccination coverage (VC) estimates were obtained from provincial and territorial immunization repositories through the Canadian COVID-19 Vaccination Coverage Surveillance System (CCVCSS). The numbers of people vaccinated were aggregated by jurisdiction, week, age group (i.e., 12–17, 18–39, 40–59, 60–79, and 80+ years) and vaccination status. Yearly population estimates were obtained from Statistics Canada and supplemented by the Northwest Territories and Yukon governments. The unvaccinated population was calculated by subtracting the number of people with at least one dose of a COVID-19 vaccine from the population estimate. For the weeks when population estimates were lower than the population with at least one dose, the latter was used as the population estimate for each jurisdiction.

This analysis covers the period of December 14, 2020, the start of the Canadian COVID-19 vaccination campaign, to January 1, 2022. By January 1, 2022, most jurisdictions had reduced the scope of their testing strategies to prioritize individuals at higher risk of experiencing severe outcomes. For incidence rate analyses, the period of June 19, 2021, to January 1, 2022, was used due to VC data availability. Data were extracted on April 28, 2023, from the National COVID-19 Dataset and on April 23, 2023, from CCVCSS.

Definitions

This analysis includes COVID-19 cases that met the national confirmed case definition ^{Footnote 11}. Vaccination statuses (defined in Table 1) were assigned using information about the number of doses received, time interval between vaccination and episode date, and vaccine product as per Health Canada authorization ^{Footnote 4}. Vaccination statuses were derived from VC definitions and include time to build immunity ^{Footnote 12}.

Data analysis

Descriptive statistics were computed to explore demographic and clinical characteristics of COVID-19 cases by vaccination status. To visualize changes in VC in Canada, the proportion of the population with a completed primary series was plotted over time by age group and a 14-day lag was applied to coverage counts to account for time to build immunity ^{Footnote 12Footnote 15}.

To contextualize changes in severity and transmissibility due to circulating variants, six variant periods were defined: wild-type, mixed variant of concern (VOC) emergence, mixed VOC predominance, Delta emergence, Delta predominance, and Omicron emergence in Canada. The start and end of a variant predominance period occurred when the specified variant first and last accounted for 75% of sequenced cases, hospitalizations, ICU admissions, and deaths. When cases, hospitalizations, ICU admissions, or deaths predominance dates were different, the latest start date was used to capture the most specific cut point for all indicators. Emergence periods were defined as the day following the end of a predominance period until the next variant reached predominance. The mixed VOC period includes the Alpha, Beta, and Gamma variants, as no single VOC represented over 75% of sequenced cases.

Incidence rates were calculated using VC data as denominators. Denominator data were not available during this analysis period for cases with a complete primary series and one additional dose; these cases were grouped with those with a complete primary series for incidence rate calculations. Population estimates were used to calculate population fractions by age group for computing weekly age-standardized incidence rates for cases, hospitalizations, and deaths. To compare trends between variant periods, the average weekly incidence rate and incidence rate ratios (IRR) of these averages were computed by vaccination status for each variant period, similar to previously published methodologies using COVID-19 surveillance data ^{Footnote 16Footnote 17}. Case data were cleaned using SQL in Metabase and were analyzed using R Statistical Software version 4.0.4.

Data quality, missing data, and reporting delays

Vaccination data were available for 12 of 13 jurisdictions (all except Québec), representing 78% of the Canadian population ^{Footnote 18}. Cases less than 12 years old were excluded since they were not eligible for vaccination until November 19, 2021 ^{Footnote 6}. Cases with missing information on vaccination status or age were excluded. Cases categorized as “not yet protected,” “primary series completed and two additional doses,” and “unknown status” were excluded. Vaccination coverage was consistently reported by jurisdictions beginning June 5, 2021; as such, incidence rates were calculated for cases with episode dates of June 19, 2021, onward (accounting for two weeks to build immunity) ^{Footnote 19}.

Results

Vaccination coverage gradually increased in Canada from December 2020 as vaccination eligibility expanded, with the proportion of individuals aged 12 years and older with a completed primary series reaching over 80% by the end of 2021 (Figure 1). From late-2020 to mid-2021, unvaccinated individuals accounted for the highest proportion of cases; however, this proportion decreased as more vaccinated individuals became cases in late 2021, aligning with the gradual increase in VC. A larger increase in the proportion of vaccinated cases was noted in December 2021, following the emergence of the Omicron variant ^{Footnote 20}.

Figure 1 - Text description

Figure 1 displays trends in vaccination coverage and COVID-19 cases and outcomes by vaccination status from December 14, 2020, to January 1, 2022. The first panel, “Vaccination coverage”, shows a line chart of the proportion of the Canadian population who completed a primary series for COVID-19 by a 20-year age group along with key vaccine rollout and variant predominance period start dates. In early 2021, coverage was close to 0% for all age groups but by the end of 2021, all age groups had over 75% coverage, with those aged 80+ years consistently having the highest coverage. The second panel, “Cases”, displays a bar plot of the weekly proportion of COVID-19 cases overlaid with a line showing the overall weekly case incidence. The third and fourth panel show the same type of plot as the second panel but for hospitalizations and deaths, respectively. All three panels show similar distribution of cases by vaccination status over time where unvaccinated individuals accounted for the overall highest proportion of cases, though this proportion decreased over time as the proportion of cases partially vaccinated and with a complete primary series increased. In early December, there was a large resurgence in case incidence which was proportionally larger than that of hospitalizations and deaths.

From December 14, 2020, to January 1, 2022, a total of 1,194,694 COVID-19 cases with complete case-level vaccination history (73.6% of all cases) over the age of 12 years were reported to PHAC (Table 2). The majority of these cases were unvaccinated, and the lowest proportion of cases was among those with a completed primary series and one additional dose. The proportion of cases was higher among females than males. Unvaccinated cases were generally younger than vaccinated cases. The highest proportions of hospitalized cases and deaths were reported among those who were unvaccinated, followed by those who completed a primary series.

During the Delta emergence period, age-standardized incidence rates for all COVID-19 outcomes remained low for all vaccination statuses (Figure 2 and Table 3). During the Delta predominance period, incidence rates increased for cases in August; severe outcomes peaked in mid-September, remaining elevated until mid-December. The increase was more pronounced among unvaccinated cases. Incidence rates for cases, hospitalizations, and deaths were consistently highest in the unvaccinated and lowest in those with a completed primary series from mid to late-2021. However, in mid-December 2021, overall cases and severe outcomes rapidly increased, and the case incidence rate among those who completed a primary series surpassed that of the unvaccinated.

Figure 2 - Text description

Figure 2 displays three line plots of the weekly age-standardized incidence rate of COVID-19 cases (top panel), hospitalizations (middle panel), and deaths (bottom panel) per 100,000 population by vaccination status from June 19, 2021, to January 1, 2022. Each plot also shows a dotted line representing the weekly incidence rate ratio of the unvaccinated incidence rate over the incidence rate for those with a complete primary series. In all three plots, the incidence rate was highest among those who were unvaccinated, followed by those who were partially vaccinated, and the lowest incidence rates were in those who completed a primary series. In late 2021, there was a large increase in cases and the incidence rate of cases with a complete primary series surpassed that of the other two vaccination status groups; however, this increase was only observed in cases while trends remained the same throughout the year for hospitalizations and deaths with the lowest rates being in those with a complete primary series.

Incidence rate ratios (IRR) for unvaccinated cases, hospitalizations, and deaths compared to those with a completed primary series remained high for most of the year (Figure 2). During the Delta emergence period, unvaccinated people were 11.4 and 17.5 times as likely to be hospitalized or to die due to COVID-19, respectively, compared to people who completed a primary series (Table 3). During the Delta predominance period, the IRR of the unvaccinated compared to those with a completed primary series increased for hospitalizations and decreased for deaths, compared to the Delta emergence period. During the Omicron emergence period, there was a decrease in IRR for cases, hospitalizations, and deaths, when compared to the Delta predominance period, though the IRR for cases had a more pronounced decrease. Following the emergence of the Omicron variant in mid-November, the weekly IRR decreased as more vaccinated people became infected (Figure 3).

Figure 3 - Text description

Figure 3 shows a 3x3 panel plot of weekly age-specific incidence rates of COVID-19 cases, hospitalizations, and deaths (3 rows) by vaccination status for unvaccinated, partially vaccinated, and primary series completed (3 columns). Each panel shows age-specific incidence rates by 20-year age group for each respective combination of outcome and vaccination status. Individuals aged 80+ years consistently had the highest incidence rates over time for both hospitalizations and deaths for all three vaccination statuses. As the number of doses increased (from unvaccinated to partially vaccinated to primary series completed), the incidence rate decreased, except for the case incidence, which experienced a large increase in late 2021.

Age-specific rates of severe outcomes were consistently highest among those aged 80 years and older, followed by those aged 60 to 79 years, for all vaccination statuses (Figure 3). Case incidence rates were highest among individuals aged 18 to 39 years, followed by those aged 40 to 59 years, from mid-2021 to late August 2021. The incidence rate of cases in these two age groups declined following VC increases until the Omicron emergence period.

Discussion

Summary of key results

From December 14, 2020, to January 1, 2022, a total of 1,194,694 cases of COVID-19 over 12 years of age and with complete case-level vaccination history were reported to PHAC. From mid to late-2021, incidence rates for cases, hospitalizations, and deaths were consistently highest among unvaccinated people and lowest among those with a completed primary series. In December 2021, following the emergence of the Omicron variant, the case incidence rate among those who completed a primary series surpassed that of the unvaccinated; however, rates of severe outcomes remained lower among those who completed a primary series.

Key results and comparison

Vaccination coverage and case incidence

Vaccination coverage steadily increased in Canada as eligibility expanded, varying by age groups ^{Footnote 3Footnote 12}. Like in the United Kingdom, case incidence in 2021 was consistently highest in unvaccinated individuals, with younger age groups having the highest incidence rates ^{Footnote 17Footnote 21}. Starting in spring 2021, there was an increase in vaccine breakthrough cases, consistent with studies showing that, although completion of a primary vaccination series was highly effective in preventing infection against the wild-type virus and Alpha variant, it was slightly less effective against the Beta, Gamma, and Delta variants ^{Footnote 22}.

Severe outcomes

Incidence rate ratio analyses showed that people who were unvaccinated were much more likely to be hospitalized and to die from COVID-19 than those who completed a primary series during the Delta emergence, Delta predominance, and Omicron emergence periods. Although IRRs for hospitalization and death decreased during the Omicron emergence period, protective effects against severe outcomes were still observed in those who completed a primary series. These trends are similar to those reported in the United States during the same period ^{Footnote 17}. Incidence rates for severe outcomes were highest in individuals aged 80 years and older, in agreement with studies showing that advanced age increased the risk of COVID-19 death ^{Footnote 16Footnote 23}. Among individuals aged 80 years and older, severe outcomes were lowest in those who completed their primary series, followed by those who were partially vaccinated, consistent with studies illustrating that a primary vaccine series was highly protective against severe outcomes from the Alpha, Beta, Gamma, Delta, and Omicron variants ^{Footnote 16Footnote 17Footnote 21}, and that partial vaccination was also effective at preventing hospitalizations and deaths ^{Footnote 15Footnote 24}.

Waning immunity during Delta period

When the more severe and transmissible Delta variant was predominant ^{Footnote 25}, there was an increase in the incidence of cases among individuals who completed a primary series, similar to trends observed in other countries ^{Footnote 17Footnote 26Footnote 27Footnote 28}. This shift could be due to potential waning of vaccine-induced immunity against symptomatic infection ^{Footnote 29Footnote 30}, a longer time since vaccination ^{Footnote 26Footnote 27}, and reduced effectiveness of available vaccines at preventing infection against the Delta variant ^{Footnote 22}. Although case IRRs decreased during the Delta predominance period, IRRs for hospitalizations and deaths remained high, suggesting that a primary series was still protective against severe outcomes, consistent with the literature ^{Footnote 17Footnote 26}. Age-stratified incidence rates for hospitalizations and deaths in fall 2021 were substantially higher among cases aged 80 years and older across all vaccination statuses. This age group was prioritized for vaccination and completed their primary series earlier in the year, further raising concerns about waning immunity ^{Footnote 26}. Routine monitoring of severe outcomes following vaccination helped inform NACI booster dose recommendations, which recommended earlier booster shots for individuals at higher risk of severe illness ^{Footnote 8}.

Emergence of the Omicron variant

The introduction of the immuno-evasive Omicron variant ^{Footnote 20} in mid-November 2021 was followed by a resurgence in cases and severe outcomes, corresponding to when over 85% of the population over the age of 12 years had completed their primary series. Although there was a substantial increase in case incidence among individuals who completed a primary series, the increases in the incidence of hospitalizations and deaths were proportionally lower than that of cases. Moreover, even with the increase in case incidence following this case resurgence, severe outcomes remained lowest among those who were vaccinated. Primary vaccination series still conferred good protection against severe outcomes from the Omicron variant, despite reduced protection against infection ^{Footnote 17Footnote 22}. Studies showed that a booster dose offered additional protection against infection and severe outcomes with the Omicron variant ^{Footnote 23Footnote 31Footnote 32Footnote 33}.

Strengths and limitations

Of 13 jurisdictions in Canada, 12 and 13 regularly reported case-level vaccination history data and VC data to PHAC, respectively. Strong participation and collaboration between federal and jurisdictional entities in Canada, alongside widespread community testing, enabled monitoring of highly representative national trends in COVID-19 outcomes following vaccination during this period. As such, the differential impacts of vaccination on COVID-19 outcomes by demographics and SARS-CoV-2 variants were effectively captured as vaccines became more widely available and administered in Canada.

Vaccination rollout timelines and uptake, testing strategies, PHMs, and VOC emergence differed across and within jurisdictions; therefore, national trends should be interpreted with caution. Denominator data were not available before June 2021 as VC data was not consistently reported from jurisdictions, precluding analysis of more stable incidence rate trends during this period. Public health testing in many jurisdictions prioritized high-risk individuals and healthcare workers during this period, which may have introduced bias into earlier descriptive trends. Additionally, distribution of rapid antigen tests to the population could lead to an underestimate of PCR-confirmed cases in late 2021. The overlap of variant periods with circulating VOCs could have introduced bias in the results. The analysis period ends shortly after the emergence of the Omicron variant and does not fully capture waning of vaccine-induced immunity. Analyses could not account for reinfection and natural or hybrid immunity, as these data were not available. This analysis does not include cases aged younger than 12 years, as they were not eligible for vaccination for most of the analysis period. Demographic data were limited to age and gender, as data on race/ethnicity and socioeconomic status were not available. Lastly, cases that were excluded from the analysis due to missing or unknown data (e.g., vaccination status) may differ from those that were included by characteristics for which data were not available (e.g., health conditions).

Conclusion

In Canada, hospitalizations and deaths due to COVID-19 were highest in older age groups; however, vaccination reduced the incidence of severe outcomes by a notable margin across all age groups. Routine monitoring of COVID-19 outcomes by vaccination status is an important pillar in Canadian COVID-19 epidemiology and surveillance to understand the impact of vaccines across Canada. It has informed Canadians about the COVID-19 epidemiologic situation in Canada and provided evidence to support policies, directives, and recommendations on vaccination and public health interventions from NACI, the PHAC Office of the Chief Public Health Officer, and jurisdictions. Though the landscape of COVID-19 is ever-changing, vaccination remains one of the most important public health interventions to protect against COVID-19 severe outcomes.

Authors' statement

• DD — Conceptualization, methodology, software, formal analysis, writing–original draft, writing–review and editing, visualization
• SM — Conceptualization, supervision, methodology, writing–original draft, writing–review and editing
• MC — Conceptualization, methodology, software, writing–review and editing
• CC — Conceptualization, methodology, writing–review and editing
• BHMF — Writing–review and editing
• FB — Writing–review and editing
• JR — Writing–review and editing
• SD — Conceptualization, supervision, writing–review and editing

Acknowledgements

The authors wish to thank provincial and territorial surveillance partners; the Vaccination Coverage and Information Systems team, including Cindy Hong, Sophia Roubos, Ahash Jeevakanthan and Donalyne-Joy Baysac; Steven Buckrell, formerly on the COVID-19 Epidemiology and Surveillance team, the Data Integration team and the COVID-19 Epidemiology and Surveillance team at the Public Health Agency of Canada.

Funding

This work was supported by the Public Health Agency of Canada as part of its core mandate.

References