Thread by @juliette_io

NSW Wild TBDs - status confirmed 🧵1/8

Each day ~30-40% of new cases are Wild TBD

Surveillance reports fill in the blanks but the data is 2-3 weeks old

Here's what they tell us ...

1. High Risk / Full Wild (daily) -

#COVID19nsw #covidnsw #COVID19Aus

2/8. These charts show data from 2 sources -

- Surveillance report (solid line)
- Daily announcements - pressers / tweets (Dashed line)

The solid line extends to 25/7 (latest date in surveillance reports)

The dashed line extends to today

2. Medium Risk / Part Wild (daily) -

3/8. Surveillance Reports categorise Wild data as follows:

- High risk
- Medium risk
- Low risk

These fit perfectly with the 3 categories we use:

- Full wild
- Part wild
- Isolation

3. Low Risk / Isolation (daily) -

4/8. What can we see?

The solid lines show us that once case interviews have been done, there are upwards revisions to all 3 categories

They're fairly evenly distributed

4. High + Medium Risk / Full + Part Wild (daily) -

5/8. So the fact that a case is Wild TBD doesn't mean it's more likely to be Wild

BUT ...

The charts above cover the period *before* Wild TBDs exploded

The chart below shows the period covered by surveillance reports so far

The next period has many more Wild TBDs -

6/8. Upshots / takeaways ...

For the time being, it seems safe / logical to think of the cases with Wild status confirmed as a fairly accurate sample

The distribution we *can* see is probably similar to the distribution down we *can't* see

7/8. For that reason, I'm going to include a sample analysis in the daily reporting

This should help us make an educated guess about the Wild TBDs

https://twitter.com/juliette_io/status/1424197501448458241

8/8. Of course, this could change

The next surveillance report will cover a period with many more Wild TBDs. When we see that, we'll know more

It should come out on Thursday