This story about Texas not including antigen results in its public reports came directly from the HC Data Team's nightly counts.

@jordan_rubio @stephanierlamm and I got the tip from updating our numbers and chased the issue from there. https://www.houstonchronicle.com/news/investigations/article/Thousands-of-Texans-are-getting-rapid-result-15452709.php

The San Antonio mayor and Bexar County judge openly talked about including "probable" cases in its totals since May.

But the county didn't separate them out in its totals.
We started asking around to see what these cases were and where they came from.

Turns out the "probable" cases weren't from past infections (antibody tests) or contact tracing. It came entirely from rapid response antigen tests.

The same tests that are increasingly popular as the response time for the nose swab PCR tests gets longer and longer.

We asked the state for a count of antigen tests since counties are sending that data to them.
Didn't have it.
We asked if they had the number of positive antigen tests.
Didn't have it.

Why did it take to mid-July for the state to realize Bexar was reporting antigen positives in its totals?

Cause Bexar didn't break it out on their COVID dashboard.

But doesn't the state get the data from the counties and private test labs directly?
Well.....
It does. But with more than 400,000 forms from PCR positives alone, the state isn't able to use those forms for counts.

So how does the state decide how many people have had the coronavirus in Texas?

It looks at county websites. Just like the Data Team at the Houston Chronicle.

11 out of 254 counties have reported some kind of antigen positive counts to the public, even if the state isn't.

Those counties have more than 11,000 antigen positives listed.

So how big is the undercount? We don't know because the state doesn't know either.

Aren't they supposed to report these results to the feds (CDC) ?

Yup.

But since the state doesn't know how many antigen tests or positives it has, it hasn't reported them to the CDC either.

On top off all that, the feds are rolling out a nursing home testing initiative entirely based on antigen testing.

But any positives from those tests won't show up in Texas' totals.

So the undercount is at least 11K+. And growing.

Proud of my team ( @jordan_rubio and @stephanierlamm) for chasing this down and being the first reporters in Texas to get a sense of just how many cases were missing because of this.