COVID-19 outbreak trends in Canada, 2021

Abstract

Background: In January 2021, the Public Health Agency of Canada launched an outbreak surveillance system, the Canadian COVID-19 Outbreak Surveillance System (CCOSS), with the goal of monitoring incidence and severity of coronavirus disease 2019 (COVID-19) outbreaks across various community settings and complementing case surveillance.

Methods: Seven provinces were included in this report; these provinces submitted weekly cumulative COVID-19 outbreak line lists to CCOSS in 2021. Data includes administrative variables (e.g. date outbreak declared, date outbreak declared over, outbreak identifier), 24 outbreak settings, and number of confirmed cases and outcomes (hospitalization, death). Descriptive analyses for COVID-19 outbreaks across Canada from January 3, 2021, to January 1, 2022, were performed examining trends over time, severity, and outbreak size.

Results: Incidence of outbreaks followed similar trends to case incidence. Outbreaks were most common in school and childcare settings (39%) and industrial/agricultural settings (21%). Outbreak size ranged from 2 to 639 cases per outbreak; the median size was four cases per outbreak. Correctional facilities had the largest median outbreak size with 18 cases per outbreak, followed by long-term care facilities with 10 cases per outbreak. During periods of high case incidence, outbreaks may be under-ascertained due to limited public health capacity, or reporting may be biased towards high-risk settings prioritized for testing. Outbreaks reported to CCOSS were dominated by jurisdictions with the largest populations.

Conclusion: The trends illustrate that COVID-19 outbreaks in 2021 were reported most frequently in community settings such as schools; however, the largest outbreaks occurred in congregate living settings. The information gathered from outbreak surveillance complemented case incidence trends and furthered understanding of COVID-19 in Canada.

Introduction

Context

On January 25, 2020, the first severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection was detected in Canada ^{Footnote 1}. The World Health Organization (WHO) declared coronavirus disease 2019 (COVID-19), the disease caused by SARS-CoV-2 infection, a pandemic on March 11, 2020 ^{Footnote 2} and COVID-19 rapidly spread across Canada with many cases and deaths associated with outbreak events ^{Footnote 3 Footnote 4}. Outbreaks can result in many infections over a short period of time, which has the potential to rapidly increase community transmission through secondary exposures and strain healthcare services ^{Footnote 4}.

The Public Health Agency of Canada (PHAC) determined that outbreak surveillance could provide additional data on the COVID-19 epidemic in Canada to help guide the public health response; as a result, PHAC created the Canadian COVID-19 Outbreak Surveillance System (CCOSS) following consultations with provincial and territorial partners. In January 2021, CCOSS launched with seven provinces contributing data: British Columbia (BC), Alberta (AB), Manitoba (MB), Ontario (ON), Québec (QC), Nova Scotia (NS) and Prince Edward Island (PE). These provinces represent 93% of the Canadian population ^{Footnote 5}. The remaining Canadian provinces and territories indicated their support for national outbreak surveillance but were unable to participate at that time. Ongoing efforts are underway to integrate interested partners.

Outbreak surveillance has not been commonly conducted for established pathogens at the national level. Very few countries, such as Ireland and the United Kingdom, reported trends on COVID-19 outbreaks ^{Footnote 6 Footnote 7}. The implementation of an event-based surveillance system to capture COVID-19 outbreak trends in Canada is novel and different from existing event-based surveillance systems in its timeliness and the level of detail captured.

Objectives

Systematic monitoring of COVID-19 outbreak trends allows for the improved understanding of settings and populations most at risk of experiencing outbreaks and of the relative impact of outbreaks on the burden of COVID-19 in Canada. This federal/provincial/territorial initiative allowed for national outbreak trends to be monitored during the pandemic. Knowledge gained from CCOSS informs public health response in Canada for prevention and control of SARS-CoV-2 transmission in various settings. This report provides a retrospective descriptive analysis of COVID-19 outbreak trends in Canada observed over a one-year period from January 2021 to January 2022.

Methods

Information sources

Data were extracted from the CCOSS database and enhanced with case-level data from the National COVID-19 Case Dataset, both stored in a Postgres database and maintained by PHAC. The CCOSS is a passive event-based surveillance system implemented in January 2021 that aims to systematically monitor the incidence and severity of COVID-19 outbreak events by setting type. Seven provinces were included in this analysis (BC, AB, MB, ON, QC, NS and PE); these provinces submitted cumulative COVID-19 outbreak line lists electronically to CCOSS on a weekly basis in 2021. The surveillance system captures aggregate outbreak-level data on administrative variables (e.g. date outbreak declared, date outbreak declared over, outbreak identifier), 24 outbreak settings and number of confirmed cases and outcomes (hospitalization, death). In 2021, provinces submitted records of outbreaks for a variety of outbreak settings, including but not limited to the following settings of interest: long-term care facilities (LTCF), acute care, school and childcare settings, correctional facilities, congregate living, and industrial settings (Table 1) ^{Footnote 8 Footnote 9}.

Table 1: Condensed grouping of outbreak settings in Canadian COVID-19 Outbreak Surveillance System

Condensed settings	CCOSS outbreak settings
Acute care	Acute care setting
Congregate living	Congregate living setting (e.g. assisted living, shelters, group homes), retirement residence
Correctional facility	Correctional facility
Industrial/agricultural	Industrial setting, agri-food processing facility
Long-term care facility	Long-term care facility
School and childcare	School, daycare, or day camp
Other	Community healthcare setting, emergency services, mass gathering event, office, personal care setting, recreational facility, restaurant/bar, retail, social event, transportation, travel/tourism, other specify, workplace unspecified, unknown
Figure 1 abbreviations Abbreviation: CCOSS, Canadian COVID-19 Outbreak Surveillance System

The National COVID-19 Case Dataset is a non-nominal case-based surveillance system capturing data on case demographics, clinical status and outcomes, exposures, risk factors, vaccination, and variant lineages of COVID-19 cases in Canada. For provinces that provided permission for linkage of outbreak and case data, CCOSS data were linked to the COVID-19 case dataset using unique outbreak identifiers. This linkage allowed PHAC to obtain additional information on outbreak case counts and severe outcomes for outbreaks with missing information and variant lineages.

Definitions

The national definition for COVID-19 outbreaks is as follows: two or more confirmed cases of COVID-19 ^{Footnote 10} epidemiologically linked to a specific setting and/or location, excluding households since household cases may not be declared or managed as an outbreak if the risk of transmission is contained. This definition also excludes cases that are geographically clustered (e.g. in a region, city, or town) but not epidemiologically linked, and cases attributed to community transmission ^{Footnote 10}.

To contextualize the impact of each variant of concern (VOC), we utilized the period from VOC introduction to the end of the dominant period. Using data from the National COVID-19 Case Dataset, the case with the earliest episode date was used to define the date of introduction. The date corresponding to the end of the dominant period was identified based on the last date that the proportion of VOC cases accounted for at least 75% of sequenced cases reported. March 2021 to May 2021 corresponded to a period of mixed VOC (Alpha, Beta, Gamma) dominance as no single VOC represented over 75%.

To provide context for the role of COVID-19 vaccines on outbreak incidence, we added a vaccination timeline for the start of the vaccination campaign in priority groups (December 2020) and the dates Health Canada authorized vaccines for individuals aged 12–17 years (May 5, 2021) and 5–11 years (November 19, 2021) and first booster doses (November 9, 2021), understanding that there are variations in the timing of the vaccine rollout in different jurisdictions ^{Footnote 11 Footnote 12 Footnote 13 Footnote 14}.

Data quality and missing data

The national COVID-19 outbreak definition was applied to CCOSS data. Denominators for outbreak settings (e.g. number of schools or populations at risk within these settings) were not available; thus, it was not possible to calculate which settings are most at risk of outbreaks from these data, only which settings most commonly report outbreaks. Therefore, calculated proportions of outbreaks by setting are relative to the overall number of outbreaks. The completeness of variables reported in the outbreak line list varied by jurisdiction. In a jurisdiction that did not report on industrial setting outbreaks separately, where possible "workplace unspecified" outbreaks were mapped to "industrial/agricultural" based on information from location names. Case fatality was defined as the proportion of confirmed COVID-19 cases associated with an outbreak that died due to COVID-19. Case fatality was missing for approximately 5% of outbreaks. Duplicate records identified using unique outbreak identifiers were removed.

Data analysis

The CCOSS outbreak settings were grouped for analyses according to Table 1. Settings that represented less than 2% of outbreaks or that were inconsistently reported across provinces were grouped into the "other" category. For reporting purposes, data were aggregated by epidemiological week (Sunday to Saturday) in which the outbreak was declared; therefore, this report covers outbreaks declared during the period of January 3, 2021, to January 1, 2022. Data were extracted on April 29, 2022. Data were cleaned and analyzed using R Statistical Software version 4.0.4. Descriptive statistics on outbreak trends by setting, over time and by case characteristics, such as outbreak size and severe outcomes, were computed.

Results

Outbreak trends by settings

From January 3, 2021, to January 1, 2022, a total of 30,078 outbreaks were reported to CCOSS from the seven contributing provinces (Table 2). Outbreaks were most common in school and childcare settings (39%), followed by industrial/agricultural settings (21%), congregate living settings (8%), LTCF (4%), acute care facilities (4%), and correctional facilities (1%). Twenty-four percent of outbreaks occurred in "other" settings. In Canada, schools are in session from January to June and September to mid-December, and during these periods, school and childcare settings consistently accounted for the largest weekly proportion of outbreaks (Figure 1). This was especially evident from September to December 2021, when children younger than 12 years of age were not yet eligible for COVID-19 vaccination; the proportion of outbreaks with vaccine-eligible populations was smaller during fall of 2021. In school and childcare settings, primary schools accounted for the largest proportion of outbreaks, followed by childcare centres, K–12 schools (combined primary and secondary schools), and secondary schools. In the "other" setting, outbreaks occurring in workplaces of an unspecified type accounted for the largest proportion of outbreaks (Figure 2). In congregate living settings, retirement residences and assisted living/group homes accounted for the largest proportion of outbreaks.

Table 2: Number and proportion of outbreaks, outbreak-associated cases, hospitalizations, deaths, and summary statistics of outbreak size by setting

Setting	Outbreaks n=30,078		Cases n=241,335		Hospitalizations n=10,252		Deaths n=3,988		Median size
	n	%	n	%	n	%	n	%	n	Range
School and childcare	11,699	39	73,311	30	294	3	7	<1	4	2–236
Other	7,069	24	39,912	17	1,031	10	157	4	3	2–364
Industrial/agricultural	6,262	21	38,777	16	939	9	103	3	3	2–300
Congregate living setting	2,508	8	34,641	14	2,960	29	1,098	28	6	2–374
Long-term care facility	1,267	4	34,703	14	885	9	1,554	39	10	2–342
Acute care	1,120	4	12,051	5	4,073	40	1,061	27	7	2–112
Correctional facility	153	1	7,940	3	70	1	8	<1	18	2–639

Figure 1 - Text description

Figure 1 displays trends in SARS-CoV-2 variants, all cases incidence, outbreak incidence by setting and proportion of outbreak setting over time. Panel A displays the period of time from when SARS-CoV-2 variants of concern were introduced to the end of their dominance in Canada, which can be contrasted against increases in case and outbreak incidence. Case (Panel B) and outbreak incidence (Panel C) peaks and troughs align over time with large increases noted in March to April 2021 and December 2021. The increase observed in December 2021 aligned with the introduction of Omicron. The proportion of outbreak setting (Panel D) illustrates that a large proportion of outbreaks were in school and childcare facilities, with a decrease in proportion noted during July and early August.

Figure 2 - Text description

Figure 2 illustrates the breakdown of other, congregate living, and school and childcare settings, to demonstrate which sub-settings had the largest proportion of outbreaks. Within the other setting, unspecified workplaces represented the largest proportion of outbreaks. Retirement residences, closely followed by assisted living/group homes represented the largest proportion of outbreaks in congregate living. Within school and childcare settings, primary schools represented the largest proportion of outbreaks.

Temporal outbreak trends

The incidence of outbreaks followed similar trends in reported cases (Figure 1). Waves denoting increasing and decreasing rates of declared outbreaks were largely driven by the introductions of VOCs and the subsequent stabilization of transmission within Canada. The increase in outbreaks in March 2021 was heavily driven by the Alpha variant, although there were also outbreaks involving Beta and Gamma variants. The Delta variant drove the resurgence in the incidence of outbreaks in August 2021, which peaked in late September 2021. The introduction of the Omicron variant in November 2021 led to a significant increase in the incidence of outbreaks but was not proportional to the magnitude of the increase observed in case incidence.

Outbreak trends by case characteristics

A total of 241,335 outbreak-associated cases were reported to CCOSS between January 3, 2021, and January 1, 2022 (Table 2). The distribution of cases by setting was similar to trends in outbreak incidence, with most cases reported in school and childcare settings (30%). Most outbreak-associated hospitalizations were associated with acute care facilities (40%) or congregate living setting (29%) outbreaks. The highest proportions of outbreak-associated deaths were in LTCF (39%), congregate living settings (28%) and acute care facilities (27%). The highest outbreak case fatality was in acute care facilities, with a mean case fatality of 10.7%, followed by LTCF (4.3%) and congregate living settings (2.9%).

The size of outbreaks ranged from 2 to 639 cases per outbreak with a median of four cases per outbreak (Table 2). Correctional facilities (18 cases per outbreak) had the highest median outbreak size followed by LTCF (10 cases per outbreak), acute care facilities (seven cases per outbreak) and congregate living settings (six cases per outbreak). Figure 3 shows that the largest outbreaks were reported in correctional facilities, with more than 18% of outbreaks reporting more than 100 cases, followed by LTCFs (6%).

Figure 3 - Text description

Figure 3 displays the proportion of outbreaks by outbreak size across settings. Correctional facilities have the largest proportion of outbreak with 100 cases or more, with most outbreaks having more than 5–9 cases. Long-term care facilities have the second largest proportion of outbreaks with 100 cases or more, while the setting "Other" has the largest proportion of outbreaks with 2–4 cases.

Condensed settings	2–4 cases	5–9 cases	10–24 cases	25–49 cases	50–99 cases	100+ cases
Correctional facility	27%	11%	20%	12%	12%	18%
Long-term care facility	32%	16%	20%	12%	14%	6%
Acute care	34%	30%	26%	8%	1%	<1%
Congregate living	38%	25%	22%	9%	4%	1%
School and childcare	60%	24%	13%	2%	1%	<1%
Industrial/agricultural	65%	21%	10%	2%	1%	<1%
Other	65%	23%	9%	2%	<1%	<1%

Within LTCF, there was a sharp decline in the incidence of outbreak-associated deaths in January 2021 shortly after the COVID-19 vaccine rollout began, for which LTCF was one of the priority settings (Figure 4). Subsequently, incidence of outbreak-associated deaths remained low and relatively stable until the introduction of the Omicron variant in late 2021 which resulted in a rapid increase in cases and deaths (Figure 1, Figure 4).

Figure 4 - Text description

Figure 4 illustrates the weekly number of long-term care facility outbreak-associated deaths over time (January 3, 2021 to January 1, 2022). Peaks in long-term care outbreak-associated deaths are observed for the first few weeks in January 2021 and in December 2021. The number of long-term care facility outbreak associated deaths is lowest during the end of May to July 2021.

Discussion

In 2021, 30,078 outbreaks were reported to CCOSS, with the highest proportions being in school and childcare, industrial/agricultural and congregate living settings. Temporal trends for outbreak and case incidence generally aligned; however, a larger relative increase was noted in cases compared to outbreaks in December 2021, following the introduction of the Omicron variant. Settings that experienced the highest severity were acute care and LTCF. The largest outbreaks were reported in correctional facilities and LTCFs, both of which are congregate living settings.

In 2021, the majority of outbreaks reported to CCOSS were in settings considered to be essential services and functions ^{Footnote 15 Footnote 16}. The most reported setting was school and childcare settings (39%). There are approximately 15,500 schools in Canada ^{Footnote 17}, representing almost 5.7 million students ^{Footnote 18} and 766,200 educators ^{Footnote 17} resulting in a large population at risk. During the Delta-driven resurgence of cases from September to November 2021, when those under 12 years of age were not yet eligible for COVID-19 vaccinations, primary schools experienced the largest increase in the incidence of outbreaks relative to other settings ^{Footnote 19}. Despite their large representation, the size of school and childcare outbreaks remained small compared to other settings, with 60% of outbreaks reporting fewer than four cases; severity in this setting remained low throughout 2021, which is consistent with what has been reported in the literature ^{Footnote 20 Footnote 21}. Studies conducted globally reported similar findings, whereby outbreak cases were small and transmission risk was low when public health measures were in place (e.g. masking, vaccination, cohorting, physical distancing, screening, testing) ^{Footnote 22 Footnote 23}.

The second most reported outbreak setting was "other," which consists of multiple settings aggregated for reporting purposes. "Unspecified workplaces" accounted for the majority of outbreaks in "other" settings. The remaining settings included restaurants, bars, retail and recreational facilities. Outbreak trends in these settings often follow case incidence trends and can contribute to community transmission given the potential for exposure from multiple networks ^{Footnote 24 Footnote 25}. However, the reported outbreaks in these settings were lower than may have been expected, potentially as a result of extended lockdown orders for nonessential settings through the course of 2021 and/or lower levels of symptom surveillance and testing compared to essential and high-risk settings ^{Footnote 25 Footnote 26}.

Although the incidence of outbreaks was highest in community-type settings (e.g. schools, retail, restaurants), outbreak size was the largest in settings with congregate living, likely due to the communal nature of these settings ^{Footnote 9}. Larger outbreak sizes may reflect increased transmissibility in high-risk populations, configuration of settings (e.g. shared bedrooms, communal areas), population dynamics (e.g. rotating staff, transient population, visitors) and outbreak management strategies (e.g. case cohorting, inability to vacate populations) within those settings. Attack rates in dormitories and shelters, both congregate settings, ranged from 1.9%–41.7% ^{Footnote 27 Footnote 28}. Denominators were not available for outbreaks submitted to CCOSS; however, high case counts in congregate settings such as retirement residences and shelters suggest attack rates may have been high.

Industrial/agricultural outbreaks accounted for the third-largest proportion of outbreaks (21%). Early in the pandemic, several large outbreaks were reported in industrial/agricultural settings (e.g. meat processing facilities, oil refineries, warehouses), many of which involved temporary foreign workers, immigrant populations and rotational workers ^{Footnote 29 Footnote 30 Footnote 31 Footnote 32 Footnote 33 Footnote 34}. Poor ventilation, difficulties maintaining physical distancing, shared transportation and housing, lack of paid sick leave and the rotational nature of the work can lead to rapid transmission of SARS-CoV-2 in certain industrial/agricultural settings—often overlapping with wider marginalized employee populations ^{Footnote 35 Footnote 36}. Additionally, many industrial and/or work camp settings involve workers from jurisdictions across Canada, which creates an elevated risk of importation of COVID-19 into the workplace, as well as back into communities in various parts of the country as workers return following rotations. Following broader vaccination rollout in May 2021, including mandatory vaccine policies in some settings, there was a dramatic reduction in the incidence of outbreaks in industrial/agricultural settings; however, this may have been in part due to revised testing strategies and outbreak management. Incidence remained relatively low until the introduction of the Omicron variant in November, which has been shown to escape immunity ^{Footnote 37}. The case fatality in this setting was 0.2%, though this could be biased by the healthy worker effect ^{Footnote 38}.

The LTCF are settings that experienced devastating impacts as a result of COVID-19 outbreaks ^{Footnote 39 Footnote 40}. In Canada, 80% of COVID-19 deaths during the first pandemic wave occurred in LTCF, when COVID-19 vaccines were not yet available ^{Footnote 40}. In November 2020, the National Advisory Committee on Immunization (NACI) recommended that the first stage of the COVID-19 vaccine rollout be prioritized to residents and staff of LTCF, among other high-risk settings such as retirement residences and acute care facilities ^{Footnote 41}. Jurisdictions across Canada reported high vaccine uptake among this population ^{Footnote 42 Footnote 43 Footnote 44 Footnote 45}. As vaccine coverage increased among the LTCF population in early 2021, there was a notable and rapid decrease in the incidence of COVID-19 deaths in this setting, aligning with evidence demonstrating vaccine protection against severe outcomes ^{Footnote 46 Footnote 47 Footnote 48}. In December 2021, we observed an increase in the incidence of deaths associated with outbreaks in LTCF, reflecting evidence of a shift in the immunity profile following the introduction of the immuno-evasive Omicron variant in fall of 2021 ^{Footnote 49}.

Outbreaks can result in high morbidity and mortality—adding strain to the healthcare system in addition to individual suffering ^{Footnote 50}. For COVID-19, the following populations have been identified as higher risk for severe outcomes: older adults (especially those living in congregate settings); people with medical conditions; pregnant people; and communities that experience disproportionate burdens of disease (People of Colour, Indigenous Peoples, refugee populations) ^{Footnote 51}. However, details on communities disproportionately impacted by COVID-19 were not available through CCOSS or case data. Within the CCOSS data, congregate living settings and LTCF experienced high case burdens; however, acute care settings had the highest case fatality (10.7%). The highest numbers of deaths were reported in LTCF, followed by congregate living settings and acute care facilities. Other countries, such as France, the United Kingdom, Belgium and Australia, have also noted severe outcomes in these high-risk populations ^{Footnote 52}. Mortality rates in LTCF have been higher than those of older adults in community dwellings ^{Footnote 53}. An international review found that 19%–72% of deaths during the pandemic occurred in LTCF ^{Footnote 52}. High mortality in acute care settings may be due to patients having comorbidities that impacted their clinical outcome ^{Footnote 54}. Approximately 90% of deaths in acute care outbreaks were among individuals older than 60 years of age ^{Footnote 54}. Interpreting severity for acute care is challenging, as individuals are already hospitalized and death rates may be influenced by underlying health conditions. Similarly, residents of LTCF who become ill have access to care in the facility rather than being hospitalized, complicating the interpretation of nonfatal severity outcomes in this setting. Continuing to monitor COVID-19 outbreaks in LTCF and acute care settings is recommended and aligns with current surveillance practices for other respiratory viruses (e.g. influenza).

Strengths

The CCOSS was rapidly and effectively implemented as a novel initiative in collaboration with federal, provincial and territorial partners in the middle of an emerging infectious disease emergency. PHAC recently conducted an evaluation of CCOSS and found that the surveillance system had good representation, with participating jurisdictions representing 93% of the Canadian population ^{Footnote 5}. Data providers consistently submitted outbreak line lists to CCOSS on a weekly basis and as a result, data were very timely. The system is flexible and was able to adapt to the needs of jurisdictions (e.g. multiple methods of data sharing, file types) and changes to data elements and format. Additionally, data were mapped to CCOSS variables by PHAC, which reduced reporting burden on provinces and improved acceptability. Outbreak data from CCOSS can also be linked to case data from the National COVID-19 Case Dataset to obtain information on outbreak-associated cases which further reduced reporting burden for certain jurisdictions, although linkage was not possible for all participating provinces. Variable completeness was excellent for the basic variables (i.e. outbreak identifier, setting, start date, number of cases) required for describing outbreak trends.

The CCOSS processes were efficient, which allowed for the timely dissemination of trends from CCOSS to various audiences in its short weekly surveillance cycle. Outbreak trend data, including outbreak setting, outbreak size and severity, have been used by the Office of the Chief Public Health Officer to provide context to the case incidence and for epidemic planning and modelling ^{Footnote 55}. The outbreak data collected from CCOSS has helped inform decisions by NACI around additional vaccine doses for LTCF residents. Epidemiologists from PHAC examined LTCF outbreak trends for several months following the implementation of the second dose of vaccine in LTCF residents (i.e. June to September 2021) to identify any increases in the incidence and size of outbreaks that could indicate waning immunity and the need for an additional vaccine dose.

Outbreaks have not been experienced equally across Canada. Smaller provinces and territories experienced fewer and less varied types of outbreaks compared with more densely populated provinces with large urban centres and/or concentrations of large industrial operations. Field epidemiologists mobilized to assist with the control and investigation of COVID-19 outbreaks in some provinces and territories, and used nationally aggregated outbreak statistics in the context of public health measures to make recommendations suitable to novel yet similar contexts.

Limitations

The accuracy and validity of the data reported to CCOSS by provinces for each outbreak setting are unknown and dependent on public health capacity to identify cases, contact trace, and establish epidemiological links. During periods of high case incidence, outbreaks may be under-ascertained due to limited public health capacity, or where high-risk settings were prioritized for testing. Outbreak size was underestimated in certain settings such as industrial settings, correctional facilities and acute care, where separate outbreaks were administratively recorded by smaller groupings for public health management (e.g. by floor, units, or another identifier). Furthermore, enhanced outbreak management implemented by some jurisdictions during periods with high surges in cases (i.e. the introduction of the Omicron variant) may impact the incidence of outbreaks reported, outbreak size and outbreak declared/end dates.

Only seven provinces of 13 provinces/territories were included in this analysis. The national outbreak analysis presented in this report was highly influenced by three of the most populous provinces in Canada (ON, QC and AB), which submitted the highest number of outbreaks (99%) and covered the greatest number of outbreak settings. Smaller participating jurisdictions and non-participating provincial and territorial jurisdictions may experience outbreak trends that differ from the national picture presented here, which was heavily influenced by the more populous provinces.

In October 2021, AB changed their outbreak definition to between two and 10 cases for certain settings ^{Footnote 56}. This would have led to an underreporting of smaller outbreaks in these settings and biased the outbreak size towards larger outbreaks.

Denominators for a number of facilities within a province or health region were not available, and even if they were, it would be difficult to identify how many facilities were affected by outbreaks since some facilities might report multiple outbreaks over time or report outbreaks by units/floors/classes, etc. The lack of denominators by setting and population at risk made it impossible to calculate attack rates for a setting or within a facility.

The data mapping of "workplace unspecified" to industrial/agricultural for jurisdictions that did not report on industrial settings separately may have contributed to non-industrial/agricultural workplaces being included. This could have contributed to an artificial increase in the number of outbreaks reported in this setting.

Conclusion

Nationally aggregated provincial outbreak data permitted trend and additional descriptive analyses to inform our understanding of where COVID-19 outbreaks occur in Canada and support ongoing efforts to reduce transmission in high-risk settings. Outbreak trends illustrated that outbreaks were reported more frequently in certain settings, specifically schools and workplaces; however, congregate living settings were prone to larger outbreaks and acute care settings experienced the highest case fatality. The information gathered from outbreak surveillance complemented case incidence trends and furthered understanding of COVID-19 in Canada. Monitoring COVID-19 outbreaks in high-risk settings such as LTCF, acute care and congregate living aligns with current surveillance practices for other respiratory viruses and continues to be a priority.

Authors' statement

DD — Conceptualization, methodology, software, formal analysis, writing–original draft, writing–review and editing, visualization

EM — Conceptualization, methodology, writing–original draft, writing–review and editing

AB — Conceptualization, writing–review and editing

CC — Conceptualization, writing–review and editing

JE — Conceptualization, writing–review and editing

RM — Conceptualization, supervision, writing–review and editing

KP — Conceptualization, supervision, writing–review and editing

Competing interests

None.

Acknowledgements

We thank the following current and former team members that contributed to the management of the Canadian COVID-19 Outbreak Surveillance System (CCOSS) and CCOSS analyses and products since March 2021: Samara David, Yuhui Xu, Carolyn Dohoo, Krista Wilkinson, Yasmina Tehami, Arzoo Alam, and Meong Jin (Ellie) Joung. We also thank the participating provinces (BC, AB, MB, ON, QC, NS, PE) for collecting and submitting their data to the Public Health Agency of Canada (PHAC) in 2021. Finally, we thank the Data Integration team at PHAC for their conceptualization of the database and ongoing support to integrate submitted data into the database.

Funding

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