Respiratory Virus Detections in Canada

Attribute Details
Source Name rvdss
Data Source Respiratory Virus Detection Surveillance System (RVDSS) from the Public Health Agency of Canada
Geographic Levels Nation, region, province, individual laboratories (see Geographic Codes)
Temporal Granularity Weekly (Epiweek) (see date formats docs)
Reporting Cadence Weekly, typically the Thursday or Friday after the end of an epiweek
Temporal Scope Start Epiweek 36 2013 (2013-09-07)
License Open Government Licence - Canada

Overview

This data source of respiratory virus detections in Canada is collected by the Respiratory Virus Detection Surveillance System (RVDSS) and published by the Public Health Agency of Canada (PHAC). Laboratory tests for various respiratory illnesses are reported on a weekly basis by sentinel laboratories across Canada to the Centre for Immunization and Respiratory Infectious Diseases (CIRID), a subsidiary of PHAC. The data was originally reported in weekly online reports, but since June 2024, has changed to being reported through a dynamic dashboard.

Note: Human coronovirus (HCoV) refers to seasonal coronovirus, which differs from SARS-CoV-2, the novel pandemic coronovirus that causes COVID-19. SARS-CoV-2 (COVID-19) was not reported until the start of the 2022-2023 season.

Table of contents

  1. Respiratory Virus Detections in Canada
    1. Signals
    2. Geography
    3. Estimation
      1. Calculation of Percent Positive Lab Tests
    4. Missingness
    5. Lag and Backfill
      1. Data Versioning
      2. Revisions
    6. Limitations
    7. The API
      1. Parameters
        1. Required
        2. Optional
      2. Response
    8. Example URLs
      1. RVDSS at the National Level for a range of epiweeks
      2. RVDSS for a specific province and specific issue
    9. Code Samples
    10. Source and Licensing
    11. Additional Resources

Signals

Signal Description
sarscov2_tests Number of SARS-CoV-2 (COVID-19) laboratory tests
Earliest Date Available: 2022-09-03
sarscov2_positive_tests Number of positive SARS-CoV-2 (COVID-19) laboratory tests
Earliest Date Available: 2022-09-03
sarscov2_pct_positive Percentage of SARS-CoV-2 (COVID-19) laboratory tests that are positive
Earliest Date Available: 2022-09-03
flu_tests Number of influenza laboratory tests. Tests for all influenza subtypes are included.
Earliest Date Available: 2013-08-31
flua_tests Number of influenza A laboratory tests. Since influenza tests normally are panels composed of subtests for influenza A and influenza B, we assume that the number of influenza A tests is always the same as the total number of influenza tests (flu_tests). This assumption has been validated on dates June 20, 2024-August 29, 2024, when the RVDSS dashboard explicitly reported numbers of total, flu A, and flu B tests. This field is maintained for convenience.
Earliest Date Available: 2013-08-31
flub_tests Number of influenza B laboratory tests. Since influenza tests normally are panels composed of subtests for influenza A and influenza B, we assume that the number of influenza B tests is always the same as the total number of influenza tests (flu_tests). This assumption has been validated on dates June 20, 2024-August 29, 2024, when the RVDSS dashboard explicitly reported numbers of total, flu A, and flu B tests. This field is maintained for convenience.
Earliest Date Available: 2013-08-31
flu_positive_tests Number of influenza laboratory tests positive for any influenza subtype.
Earliest Date Available: 2013-08-31
flu_pct_positive Percentage of influenza laboratory tests that are positive for any influenza subtype.
Earliest Date Available: 2013-08-31
fluah1n1pdm09_positive_tests Number of influenza laboratory tests positive for influenza A(H1N1)pdm09.
Earliest Date Available: 2013-08-31
fluah3_positive_tests Number of influenza laboratory tests positive for influenza A(H3N2).
Earliest Date Available: 2013-08-31
fluauns_positive_tests Number of influenza laboratory tests positive for influenza A (unsubtyped).
Earliest Date Available: 2016-09-03
flua_positive_tests Number of influenza laboratory tests positive for any influenza A subtype. This is the sum of influenza A(H1N1)pdm09 (fluah1n1pdm09_positive_tests), A(H3N2) (fluah3_positive_tests), and other influenza A (unsubtyped) (fluauns_positive_tests) positive tests.
Earliest Date Available: 2013-08-31
flua_pct_positive Percentage of influenza laboratory tests that are positive for any influenza A subtype
Earliest Date Available: 2013-08-31
flub_positive_tests Number of influenza laboratory tests positive for influenza B.
Earliest Date Available: 2013-08-31
flub_pct_positive Percentage of influenza laboratory tests positive for influenza B.
Earliest Date Available: 2013-08-31
rsv_tests Number of respiratory syncytial virus (RSV) laboratory tests
Earliest Date Available: 2013-08-31
rsv_positive_tests Number of positive RSV laboratory tests
Earliest Date Available: 2013-08-31
rsv_pct_positive Percentage of RSV laboratory tests that are positive
Earliest Date Available: 2013-08-31
hpiv_tests Number of human parainfluenza virus (HPIV) laboratory tests. Tests for all HPIV subtypes are included.
Earliest Date Available: 2013-08-31
hpiv1_positive_tests Number of HPIV laboratory tests positive for HPIV-1
Earliest Date Available: 2013-08-31
hpiv2_positive_tests Number of HPIV laboratory tests positive for HPIV-2
Earliest Date Available: 2013-08-31
hpiv3_positive_tests Number of HPIV laboratory tests positive for HPIV-3
Earliest Date Available: 2013-08-31
hpiv4_positive_tests Number of HPIV laboratory tests positive for HPIV-4
Earliest Date Available: 2013-08-31
hpivother_positive_tests Number of HPIV laboratory tests positive for any other HPIV subtype.
Earliest Date Available: 2013-08-31
hpiv_positive_tests Number of HPIV laboratory tests positive for any HPIV subtype. This is the sum of HPIV-1 (hpiv1_positive_tests), HPIV-2 (hpiv2_positive_tests), HPIV-3 (hpiv3_positive_tests), HPIV-4 (hpiv4_positive_tests), and other HPIV subtypes (hpivother_positive_tests) positive tests.
Earliest Date Available: 2013-08-31
hpiv_pct_positive Percentage of HPIV laboratory tests that are positive for any HPIV subtype.
Earliest Date Available: 2013-08-31
adv_tests Number of adenovirus (ADV) laboratory tests
Earliest Date Available: 2013-08-31
adv_positive_tests Number of positive ADV laboratory tests
Earliest Date Available: 2013-08-31
adv_pct_positive Percentage of ADV laboratory tests that are positive
Earliest Date Available: 2013-08-31
hmpv_tests Number of human metapneumovirus (hMPV) laboratory tests
Earliest Date Available: 2013-08-31
hmpv_positive_tests Number of positive hMPV laboratory tests
Earliest Date Available: 2013-08-31
hmpv_pct_positive Percentage of hMPV laboratory tests that are positive
Earliest Date Available: 2013-08-31
evrv_tests Number of enterovirus/rhinovirus (EV/RV) laboratory tests. Enterovirus and rhinovirus are both types of enteroviruses and according to the US CDC, “[m]ost diagnostic assays are not able to distinguish rhinoviruses from enteroviruses and have a combined single target for both viruses”.
Earliest Date Available: 2013-08-31
evrv_positive_tests Number of positive EV/RV laboratory tests
Earliest Date Available: 2013-08-31
evrv_pct_positive Percentage of EV/RV laboratory tests that are positive
Earliest Date Available: 2013-08-31
hcov_tests Number of human coronavirus (HCoV) laboratory tests.
Earliest Date Available: 2013-08-31
hcov_positive_tests Number of positive HCoV laboratory tests
Earliest Date Available: 2013-08-31
hcov_pct_positive Percentage of HCoV laboratory tests that are positive
Earliest Date Available: 2013-08-31

Geography

This data source reports data for Canada at the nation, region, province, and lab levels. For a complete list of valid geographic codes for each type, see the Canada section in Geographic Codes.

Estimation

Calculation of Percent Positive Lab Tests

For each respiratory virus, the number of total lab tests performed and number of positive lab tests are reported. For convenience, we calculate the percentage of positive tests for a given virus as:

\[\text{Percent Positive Lab Tests} = \frac{\text{Number of Positive Tests}}{\text{Number of Total Tests}} \times 100\]

Percent positive lab tests ranges from 0-100%. For HPIV, total HPIV positive tests are not always reported, so we manually calculate them by summing up the positive tests for all HPIV subtypes:

\[\text{HPIV Positive Tests} = \sum\text{HPIV}_\text{subtype}\]

for subtype = 1, 2, 3, 4, and other

Prior to the end of the 2023-2024 season, total tests and percent positivity were reported, but number of positive tests was not explicitly reported. For convenience, we manually calculated number of positive tests when only percent positive lab tests and number of lab tests were available.

\[\text{Number Positive Lab Tests} = \frac{\text{Percent Positive Lab Tests}}{100} \times \text{Number of Total Tests}\]

These values are not rounded, so some are not integers.

Missingness

The source uses multiple terms used to denote missing data:

  • NC : Not collected
  • NR : Data not reported for current week
  • NA : Not available
  • Not tested

We were unable to find detailed guidance about the difference between these categories. For convenience, we treat all these as missing (NA), but there may be subtle differences.

For epiweeks 5 and 47 of the 2019-2020 season, the reports are empty, so data from these weeks are missing.

Lag and Backfill

Data is reported about 5 days after the end of an epiweek. For June-August 2025, the dashboard containing the data was updated every two weeks, so the lag during that period was larger.

Data Versioning

Epiweeks end on Saturday, and the data is usually updated the following Thursday or Friday. The dashboard where data is currently published states when it was last updated, so we use that date as the version date.

The dashboard has only been used to publish data since June 2024. Prior to that, data was published weekly in blog post format. To make this older data more useful, we reconstructed the source’s version history, deriving the version date from metadata included in the weekly report webpages.

We defined the version date of a particular data issue to be the date the report webpage was last modified. Historically, page-last-modified dates were usually 5 days after the end of the epiweek data was being reported for (“reference date”). Occasionally, page-last-modified dates were later than that – sometimes much later, up to a year after the reference date.

Our understanding is that data in weekly reports were not modified after being published. Thus, we expect page-last-modified dates falling more than about a week after the reference date to indicate changes in the text on the page rather than changes to the data. With that in mind, if the page-last-modified date was more than 14 days after the reference date, we set the version date to be 5 days after the end of epiweek being reported. This matches the more common behavior seen.

Revisions

The data experiences backfill as counts are finalized in subsequent weeks. The amount and impact of revisions depends on the virus.

Across all indicators, most (99%) observations have less than 3 revisions. hpiv1_*,hpiv2_*,hpiv3_*, and hpiv4_* indicators have revisions even more rarely.

Across all indicators, the top 10% of observations (unique location-date pairs) with the largest relative spread have had values revised by 25-33% or more. However, revisions that greatly impact a reported value happen soon after an observation is first reported. 90% of observations with revisions are revised to within 20% of their final value (and stay within that 20% range) 1-3 weeks after the date an observation is being reported for. And 95% of observations with revisions are finalized within 3-4 weeks.

Looking at the revision speed of specific indicators, sarscov2_* and flub_* indicators are finalized fairly fast (95% finalized after 12 days). fluah3_positive_tests and adv_* indicators are finalized fairly slowly (95% finalized after 30-40 days).

flu_pct_positive, flua_pct_positive, fluah1n1pdm09_positive_tests and related indicators have larger spread than other indicators. That is, about 70-90% of observations have more than 10% relative spread vs 20-30% in other indicators.

Limitations

This data reports the results of laboratory testing of respiratory viruses, which only represents a subset of people who may be sick. People with mild symptoms may be excluded, and there may be a delay between someone developing symptoms and getting a lab test.

Data comes from labs across Canada, and some provinces only have one reporting lab, with no further information about, for example, location of individual labs, if any. Only Quebec and Ontario report detailed disaggregated (lab-level) data. Other provinces may have multiple labs, but don’t report them separately. If labs or healthcare providers tend to be located in larger cities or cities in general, and it is easier for people in more densely populated areas to have lab tests done than those in rural locations, the data may overrepresent the urban population.

Our reconstruction of version dates prior to June 2024 may be based on incorrect assumptions.

The API

The base URL is: https://api.delphi.cmu.edu/epidata/rvdss/

Parameters

Required

Key Description Type
geo_values The geographical location (see Geography) string
geo_type The type of geographical location (see Geography)
Available types: lab,province, region, nation		 string
time_values The date corresponding to the last day (Saturday) of epiweek date
time_type Type of time value, only week available string

Optional

Parameter Description Type
as_of issue date of the data (see Data Versioning). Mutually exclusive with issues. epiweek
issues list of issue dates (see Date Formats). Mutually exclusive with as_of. list of epiweeks

Response

Field Description Type
result result code: 1 = success, 2 = too many results, -2 = no results integer
epidata list of results array of objects
epidata[].geo_type geographical location type (nation, region, province, lab) string
epidata[].geo_value geographical location identifier string
epidata[].region region identifier string
epidata[].time_type type of time value (week) string
epidata[].epiweek epiweek for which data is reported (YYYYWW) integer
epidata[].time_value date corresponding to the last day (Saturday) of the epiweek integer
epidata[].issue issue/version date of the data integer
epidata[].year year for which data is reported (YYYY) integer
epidata[].* various signal fields (see Signals) float
message success or error message string

Example URLs

RVDSS at the National Level for a range of epiweeks

https://api.delphi.cmu.edu/epidata/rvdss/?geo_type=nation&geo_values=ca&time_values=202301-202310

RVDSS for a specific province and specific issue

https://api.delphi.cmu.edu/epidata/rvdss/?geo_type=province&geo_values=on&time_values=202340&as_of=202341

Code Samples

Libraries are available for R and Python.

from epidatpy import EpiDataContext, EpiRange

epidata = EpiDataContext()
res = epidata.pub_rvdss(
    geo_type="nation",
    geo_values="ca",
    time_values=EpiRange(202301, 202310)
)
print(res.df())

library(epidatr)

res <- pub_rvdss(
    geo_type = "nation",
    geo_values = "ca",
    time_values = epirange(202301, 202310)
)
print(res)

Source and Licensing

This source is derived from PHAC’s Respiratory Virus Detection Data, originally reported in weekly reports, and currently reported in a dynamic dashboard.

The data is made available under the Open Government Licence - Canada.

Additional Resources

PHAC provides a more detailed overview of respiratory surveillance in Canada.