The stats for COVID in Texas for the last two weeks (ending October 31) were dominated by outbreaks in El Paso and along the US 287 corridor from Ft. Worth northward.  The stats for the rest of the State changed little from the previous two weeks.

Hospitalizations

Total hospitalizations and ICU bed usage rose sharply over the last two weeks.  Total hospitalizations rose from 4,275 to 5,685 (+33%).  ICU bed usage increased from 1,290 to 1,631 (+26%).  However, approximately 70% of the increase came from El Paso and, to a lesser degree, from cities along the US 287 corridor from Fort Worth to the Panhandle.  The increase in El Paso is stunning.  Its hospitalizations jumped 128% in the last two weeks.  Notwithstanding that El Paso is only 2% of the State’s population, it now has 16% of the State’s total hospitalizations.  This chart demonstrates just how sharply El Paso’s hospitalizations have increased.

Hospitalizations have also increased significantly in communities in north Texas, generally along the US 287 corridor.  That region has about 10% of the State’s population but contributed 30% of the increase over the last two weeks and have about double its share of current hospitalizations.

In the rest of the State, hospitalizations rose far less at 17%.  Two of the State’s 22 reporting regions had a decrease in hospitalizations.  In the Houston region we have been vacillating in the 500-700 range for the last two months, which is down by 78% from the high point in July.

This is the chart for all hospitalizations in the State since the pandemic began.  There are two things that should be noted.  First, total hospitalizations, even with the current run-up, are barely over half of what they were in July.  Second, the slope of the increase (as indicated by the red arrows) is much less steep than it was in the first wave.  These are indications that the ratio of hospitalizations to new positive tests results is declining.  I will be writing more about this phenomenon shortly.

Fatalities

Average daily reported fatalities ticked up slightly from 71 to 74 but were accelerating slightly more rapidly toward the end of the two-week period.

Many are predicting a significant increase in fatalities over the winter.  However, so far, we do not see that in the data.  And while some increase seems likely to be inevitable with the rising hospitalizations, there is some reason for hope because the data is also showing an increasing improvement in the ratio of fatalities to new positive test results and hospitalizations.

Testing Results

Daily test results for the two-week period increased by about 10% from the previous two weeks but positive results increased by 30%, suggesting that about a third of the increase came from additional testing and about two-thirds from rising cases.  Generally, the increase in positive test results came from the same areas where hospitalizations increased.  The positivity rate based on the daily reports rose from 7.1% to 8.4%.  The State’s report, based on the actual specimen collection date, also increased from 8.6% to 9.8%.

Just to reiterate, I do not have much faith in test results as a reliable indicator and I refuse to refer to new positive test results as “new cases.”  Testing results from Texas have been particularly erratic.

Conclusion

The variation in COVID in different regions of the State, underscores the degree to which outbreaks are localized.  Also, it shows how COVID moves through communities in waves.  El Paso had been largely spared a serious outbreak prior to the last month.  In the meantime, previous hotspots like the Valley and Houston have been relatively quiet.

One of the underlying assumptions of a winter surge is that lower temperatures cause people to stay indoors more.  The surges in Texas certainly do not support that theory.  The surges in the Valley and Houston came during summer, and El Paso and north Texas were both unseasonably warm until the last week of October.  Once again, the data continues to confound all of the simplistic explanations of how and where COVID spreads.