SARS-CoV-2 variants: National definitions, designations and public health actions

Last updated: August 22, 2024

On this page

Overview

The Public Health Agency of Canada established the Federal SARS-CoV-2 Variant Surveillance Group (FSVSG) to monitor and assess the impact of variants of SARS-CoV-2 on viral transmissibility, disease severity, and efficacy of vaccines, therapeutics, and diagnostics. The FSVSG has developed national definitions, designations, and public health actions for SARS-CoV-2 Variants under Monitoring (VUM), Variants of Interest (VOI), Variants of Concern (VOC), and De-escalated Variants. Descendant lineages inherit the classification of the parent lineage unless otherwise designated. Note that the VUM and VOI designation relates to surveillance status and is not a statement of the level of concern associated with a given variant.

As of March 2023, Greek letters will only be assigned to VOCs, while VOIs and VUMs will be referred to using established scientific nomenclature systems. There are no current VOCs in Canada because Omicron has moved to the "de-escalation" category.

Variant of concern

In general, variants of concern (VOCs) are expected to have a meaningful impact on population level outcomes beyond what is happening with the current lineages.

Definition

A SARS-CoV-2 variant is a VOC if, through a comparative assessment, it has been demonstrated to be associated with one or more of the following:

or

is otherwise assessed to be a VOC by the World Health Organization (WHO);

or

is otherwise assessed to be a VOC by the FSVSG.

Actions

If a variant is determined to be a VOC, actions may include the following:

Current variants of concern in Canada
WHO label Pango lineageTable 1 Footnote * Designation date
N/A N/A N/A
Table 1 Footnote *

Includes all descendant lineages, unless otherwise designated.

Return to table 1 footnote * referrer

No current circulating lineages are designated VOCs in Canada.

Variant of interest

In general, variants of interest (VOIs) have the potential to replace the current dominant circulating lineage but the impact on population level outcomes is either unknown or not expected to be meaningfully different than current lineages.

Definition

A SARS-CoV-2 variant is a VOI if it:

has a genome with mutations associated with changes in epidemiology, antigenicity, or virulence, or changes that potentially have a negative impact on available diagnostics, vaccines, therapeutics, or public health measures;

and

is known to have a growth advantage over the currently circulating lineages but the population impact is unknown or expected to be similar to the currently circulating lineages in Canada or internationally;

or

is otherwise assessed to be a VOI by WHO;

or

is otherwise assessed to be a VOI by the FSVSG.

Actions

If a variant is determined to be a VOI, actions may include the following:

Current variants of interest in Canada
WHO label Pango lineage Designation date
Omicron BA.2.86 (excluding JN.1) Table 2 Footnote * December 1, 2023
Omicron JN.1Table 2 Footnote * December 21, 2023
Table 2 Footnote *

Includes all descendant lineages, unless otherwise designated.

Return to table 2 footnote * referrer

Variant under monitoring

In general, variants under monitoring (VUM) have interesting genomic or epidemiological traits but their ability to establish at the population level in Canada is unknown or not expected.

Definition

While all SARS-CoV-2 variants are under monitoring in Canada, a variant is specifically a VUM if it has genetic changes with potential to affect virus characteristics or has early signals of growth advantage relative to other circulating variants, but for which current evidence is limited or unclear requiring enhanced monitoring and reassessment pending new evidence.

Actions

If a variant is determined to be a VUM, actions may include the following:

Current variants under monitoring in Canada
WHO label Pango lineage Designation date
Omicron KP.3.1.1 Table 3 Footnote * August 22, 2024
Table 3 Footnote *

Includes all descendant lineages, unless otherwise designated.

Return to table 3 footnote * referrer

De-escalation

The SARS-CoV-2 virus continuously evolves, generating new lineages over time. Some lineages may become prevalent while others will become extinct and no longer pose a threat to public health. Therefore, lineages can be escalated from VUM to VOI to VOC but can also be de-escalated as new lineages emerge and replace currently circulating lineages.

VOI or VUM de-escalation is initiated following internal evidence review by FSVSG. A VOC is considered for de-escalation by FSVSG when any of the following are true:

Variants of concern and variants of interest that have been de-escalated after review by the FSVSG
WHO label Pango lineageTable 4 Footnote * De-escalation date
Alpha B.1.1.7 August 18, 2022
Beta B.1.351 August 18, 2022
Gamma P.1 August 18, 2022
Delta B.1.617.2 August 18, 2022
Omicron B.1.1.529 August 9, 2023
Epsilon
  • B.1.427
  • B.1.429
November 12, 2021
Zeta P.2 September 09, 2021
Eta B.1.525 March 17, 2022
Theta P.3 December 09, 2021
Iota B.1.526 December 09, 2021
Kappa B.1.617.1 October 01, 2021
Lambda C.37 December 09, 2021
Mu B.1.621 February 17, 2022
N/A A.23.1 October 14, 2021
N/A B.1.1.318 January 27, 2022
N/A B.1.617.3 July 20, 2021
Omicron XBB.1.5Table 4 Footnote * March 7, 2024
Omicron EG.5Table 4 Footnote * March 7, 2024
Table 4 Footnote *

Includes all descendant lineages, unless otherwise designated.

Return to table 4 footnote * referrer

Weekly variants breakdown and nowcasting

The weekly variant breakdown has the latest updates about variants in Canada. This graphical view is based on whole genome sequencing by federal and provincial laboratories across Canada.

In August 2023, the Public Health Agency of Canada (PHAC) began showing two weeks of nowcasted predictions on the COVID-19 variants in Canada online graph, instead of just showing a percentage mix of COVID-19 variants detected in Canada with a three week delay.

Nowcasting uses statistical models to estimate the current situation given imperfect data by learning from prior trends: they do not forecast the future. Nowcast modelling is common in other domains, such as weather forecasting and economics. These mathematical methods have recently been applied to nowcasting infectious disease trends. For genomic surveillance, a nowcast model fits growth curves of the viral lineages and provides proportion estimates for periods with accumulating and incomplete data.

PHAC's National Microbiology Laboratory (NML) has developed a nowcasting model to estimate which lineages are currently growing or shrinking in prevalence in Canada. This method allows for more up-to-date estimates of the current situation of how many Canadian cases are likely to be experiencing COVID-19 infections with certain variants. These variant trends are used to inform healthcare professionals, public health authorities, and decision-makers. Unusual variants or fast-growing variants are detected and tracked in order to inform policy that protects the health of Canadians.

Accuracy of nowcasting

Before training the nowcasting model on the data, two steps are run to improve the accuracy of prediction. First, all lineages that have reached at least 50 detections in any single week within the full 10-week window are chosen for modelling, and the rest are grouped into the "other" category. Next, the selected lineages —termed sub-lineages— are reassessed to be grouped under umbrella parental lineages, in which several lineages share direct ancestry (represented in as the main colours, such as blue). Any lineages with few detections and not growing quickly will be grouped with related lineages, in order to obtain a maximum of five sub-lineages per group.

Finally, a statistical model fits the growth curves of the weekly lineage proportions for the eight weeks prior to the window for which a nowcast projection is needed. These growth curves estimate how the variants are increasing or decreasing relative to one another. The nowcast projection is the extension of these curves into the two weeks that are still accumulating data. The result is an estimate of the current set of proportions for the variants circulating in Canada, and their 95% prediction intervals, a measure of uncertainty.

Risk assessments

Page details

Date modified: