Our partners at Health Canada have requested additional metrics from the COVID Alert server.
The outcome will be Google Sheets updated daily with the server metrics data segmented by province.
Context
The Heath Canada task force will use the data that we provide as a basis for
estimating how many users are using the new exposure window functionality in the app;
to evaluate how effectively the app is breaking the chain of infection.
Context from HC: Once the new exposure window feature is live, it will enable us to estimate how many users are taking advantage of the feature (and inputting a test date or symptom date), and how many choose to use the original 14 day default. This will let us determine how effective the new feature is at reducing burden on the healthcare system and reducing “overnotifying” where someone is notified even if they were not exposed to someone during the infectious period.
In addition, for users that do input a symptom onset date, it will allow us to calculate the average days between symptom onset/test date and a user uploading their TEKs. According to recent evidence this is likely one of the most important metrics for understanding the healthcare outcomes of the app, as the less time between symptom onset/test date and the uploading of their TEKs (and therefore notifications going out and their contacts beginning to self-isolate) the more effective the app is at breaking the chain of infection (since their contacts are self-isolating earlier). And so we plan to include this metric in our performance measurement plan.
Acceptance Criteria
Data is available by 9:00 am each day, with numbers for the previous day.
Data needs to be accurate.
The data needs to be in a Google Sheet and in format that it can easily be viewed and added into dashboards.
The data should be back-filled so that the numbers are complete. (is this possible?)
The numbers are not publically available, so need to be secured.
The data needs to be anonymous, not revealing any information about the users.
Data to track
We need to track each occurrence of a user uploading TEKs for the first time. Each record should include:
the date that the TEKs were uploaded
the province/territory where the OTK was generated
the number of TEKs uploaded
New metrics
The number of users who uploaded fewer than 14 keys vs the number of users who upload 14 keys on the first upload. (segmented by province, aggregated per day)
The average number of keys uploaded by a user on the first upload, excluding users who upload 14 keys. (segmented by province, aggregated per day/week/month)
Solution Considerations
This data is formatted as individual records so will require a database table.
We can identify if it is the first time upload by the number of keys remaining to upload in the certificate.
Our partners at Health Canada have requested additional metrics from the COVID Alert server.
The outcome will be Google Sheets updated daily with the server metrics data segmented by province.
Context The Heath Canada task force will use the data that we provide as a basis for
Context from HC: Once the new exposure window feature is live, it will enable us to estimate how many users are taking advantage of the feature (and inputting a test date or symptom date), and how many choose to use the original 14 day default. This will let us determine how effective the new feature is at reducing burden on the healthcare system and reducing “overnotifying” where someone is notified even if they were not exposed to someone during the infectious period.
In addition, for users that do input a symptom onset date, it will allow us to calculate the average days between symptom onset/test date and a user uploading their TEKs. According to recent evidence this is likely one of the most important metrics for understanding the healthcare outcomes of the app, as the less time between symptom onset/test date and the uploading of their TEKs (and therefore notifications going out and their contacts beginning to self-isolate) the more effective the app is at breaking the chain of infection (since their contacts are self-isolating earlier). And so we plan to include this metric in our performance measurement plan.
Acceptance Criteria
Data to track We need to track each occurrence of a user uploading TEKs for the first time. Each record should include:
New metrics
Solution Considerations