I've been asked on a number of occasions why I don't report on case numbers (UPDATE - I do now report on cases for Nova Scotia due to popular demand, but this post remains accurate). This post is dedicated to explaining why:

For data to be of scientific or mathematical value, there has to be context. In the case of COVID-19, context means being able to track trends and compare data from one place to the next. Trends around COVID-19 case numbers cannot be tracked with any accuracy or meaning, nor can the data be compared, here is why:

Trends - for a trend to be a trend, the data behind it has to have a standard. None exists with case numbers. This is because the testing is not random, the quantity of tests being done drastically changes from day to day and the criteria for who to tested is ever changing, not only in NS, but all over the world.

Comparability - Due to the fact the criteria of testing is ever changing, and is completely different everywhere on Earth, there is no way to competently compare case numbers from one place to the next.

As a result of these practices, the number of cases really doesn't tell a story about anything. More cases doesn't mean things are bad or worse than somewhere else, it just means more testing was done, or the criteria for testing has changed etc... The only way case numbers would be a particularly useful figure would be if:

  1. those tested were done randomly
  2. the number of tests (per capita) was the same daily.
    For example, if NS tested 1000 random people every day, you could then track the trends and know if things were getting better or worse. You still wouldn't be able to compare it to other places unless everywhere did the same randomization, but the trends within the location would be meaningful.

Ultimately, the deaths are what we are trying to minimize and that's what I'm tracking in a way that trends can be measured and compared across locations so case data is really not required anyway.

Hope that helps!

UPDATE April 19 - Also note, NS has the highest rate of testing in Canada, and one of the highest in the world, so the fact we don't have significantly more cases reported than we do is a very good sign, especially with the situation in our long term care facilities.

April 20 UPDATE