I am trying to find the accurate number of COVID cases in Harris County.
The Harris County Dashboard shows a total of 11,131 cases in Harris County.
The State Dashboard shows a total of 30,729 in Harris County.
Which is true?

Harris County COVID Dashboard:
https://harriscounty.maps.arcgis.com/apps/opsdashboard/index.html#/c0de71f8ea484b85bb5efcb7c07c6914
Note: make sure you select Harris County in top right, not Combined Cases or all of Houston.



State of Texas COVID Dashboard:
https://txdshs.maps.arcgis.com/apps/opsdashboard/index.html#/ed483ecd702b4298ab01e8b9cafc8b83

You want the combined. Harris Co is all non City of Houston cases. It' odd, but a function of what entity gathers the info.

Thanks, that is odd. The numbers are closer now at around 31,000. The daily new cases trend show that Houston/Harris county COVID cases are trending back down. Is this data correct or is there a lag in reporting?

I understand hospitalizations and deaths lag cases, which is why there talk of ICU capacity in Houston right now. But if the cases are trending back down, why declare a state of disaster in the city? The main "declaration" or "law" is the extension of the mask order, which is to prevent future cases. The cases are already trending back down to manageable levels.

State of Disaster Declaration:
https://abc13.com/harris-county-judge-lina-hidalgo-coronavirus-self-quarantine-stay-at-home-order/6281792/

It's a lag...cases on the dashboard are entered as of the date of the test. Since all the labs are 5+ days behind the most recent numbers are likely low.

Ok, thanks. I wasn't sure if a new daily case was based on the date the result of the test came back or if it was backlogged to the date the patient took the test.

It will be interesting to see if the numbers roll over or continue to increase.

I do think they report total cases each day, not just the one previous. So if there are 300 new cases in a day, that may be entered across 2 weeks on this graph but still represents the total d-o-d increase.

It would be helpful if they disclosed that on the dashboard wouldn't it? I only caught that detail by listening to a recent CoH press conference.

I am a big believer on data points need to be made public. Everything...avg age of cases, avg age of hospital admits with and without pre-existing conditions, etc. an informed public can make better decisions.

The Harris Health dashboard back dates positive test results to the day the test occurred, rather than the day the result was reported. It's a better way to look at the data, but because of the leg work involved in backdating the data is constantly being revised and so the curve shifts through time. The tail end of the chart really isn't a reliable indicator because it swings pretty dramatically.

The below chart is a composite of the same Harris Health data, pulled on different dates through the last couple of weeks, and shows how the curve gets revised relative to prior publications and the daily case data that's reported.

The Texas and Harris County dashboards both have around 31,000 cases but the daily graphs are significantly different.


https://tabexternal.dshs.texas.gov/t/THD/views/COVIDExternalQC/COVIDTrends?%3AisGuestRedirectFromVizportal=y&%3Aembed=y


https://harriscounty.maps.arcgis.com/apps/opsdashboard/index.html#/c0de71f8ea484b85bb5efcb7c07c6914

Taking the two graphs and overlaying the axes tell two very different stories. I know it's blurry. One is a peak in mid June, then starting to roll over. The other is a peak in late June that has not shown to roll over. The Harris county shows the max daily new cases at 835 on 6/11. The Texas county dashboard shows the peak at 2,000 cases on 6/23, 2 weeks later.The total number of cases is the same for both.

Fitch said:

---

The Harris Health dashboard back dates positive test results to the day the test occurred, rather than the day the result was reported. It's a better way to look at the data, but because of the leg work involved in backdating the data is constantly being revised and so the curve shifts through time. The tail end of the chart really isn't a reliable indicator because it swings pretty dramatically.

The below chart is a composite of the same Harris Health data, pulled on different dates through the last couple of weeks, and shows how the curve gets revised relative to prior publications and the daily case data that's reported.

---

Thanks, where did you get that graph? Public data or do you update it yourself?

Self made, data sourced from the Texas DSHS (raw counts) and the Harris Health website (backdated counts).

The Texas Medical Center is using the case counts from the Texas Health and Human Services Dashboard which is showing the 30,000+ figure.
https://txdshs.maps.arcgis.com/apps/opsdashboard/index.html#/ed483ecd702b4298ab01e8b9cafc8b83