Graphs updated 5/17/20
This past week I was asked to give a presentation on the epidemiology of COVID-19 in Texas to the Texas Osteopathic Medical Association. The next day, friends on social media were asking if Texas was really ready to re-open (not that it matters that much, since the governor ordered re-openings starting May 1). Found myself sharing my slides repeatedly. So here’s a summary.
In March and early April, case numbers were still doubling rapidly. The number of cases have started to decline in the past week, but we’re still seeing more cases per day than we were two months ago.
We are still well above the epidemic threshold for pneumonia, influenza, and COVID-19 like deaths. In early April, we saw more than three times the expected number of deaths due to pneumonia-like illnesses, mainly COVID-19. While the percentage has fallen to 10.6% as of May 7, in a “normal” year, these deaths would have fallen to less than 6% of all deaths by May.
While cases seemed to taper off a bit in April, the overall trend is still an increase in daily cases of COVID-19. Note that these data reflect the date the report was received* rather than the date of onset. If a lab result takes 3–5 days, the actual case date could be 5+ days prior. And the incubation period is anywhere from 2–14 days, so the current cases may or may not be due to new infections since May 1. There’s no doubt they’re increasing.
The number of deaths in Texas have fluctuated quite a bit. While we started with a 1.3% Case-Fatality Rate, as expected that has increased over the past two months to about 2.8%. This CFR is mostly consistent with what we’ve seen worldwide. I hesitate to say *any* case fatality rate that isn’t zero is “good,” but the fact it’s consistent means nothing unusual is happening in Texas cases. But remember, since late March, most Texas cities have been under Stay-at-Home orders, even if there wasn’t one statewide. Most hospitals have stopped all elective procedures and focused on COVID-19 cases. This means that hospitals have not become overwhelmed, probably thanks to both those efforts.
Bexar County and San Antonio
Here the story is a little more hopeful. First, note that these data indicate the *Date of Onset.* That means that though the cases may get reported a week later, San Antonio Metro Health epidemiologists go back into the data and update all positives to reflect the day they said they became sick. This matters, because it gives us a hint into when they were exposed. So even if it takes two weeks to receive a lab result, we can say that exposure happened 3 weeks ago. While Bexar County saw a rapid increase in cases in late March, and stable the beginning of April, new infections have started to decline in the last few weeks. There have been blips, most likely related to the jail outbreak.
A note on jails: they do not have a static population. In addition to staff going in and out, new inmates come in and others leave on a daily basis. This means new infections can come into the jail and spread to other inmates. However, while the infections came from the community, transmission inside the jail does not reflect community transmission. Having 6 inmates in a small, confined space, means the likelihood of transmission is MUCH higher than it would be for two next door neighbors who happen to be at HEB at the same time, but 6 feet apart.
Bexar County can accurately say we are trending in the right direction. But also remember we have had a Stay at Home order since March 18. We’ve limited the likelihood of exposure, limiting new cases. As people go back to work and gather, we can expect to se an uptick in cases. The only way to control this is an extensive testing, tracing, and tracking effort. The last thing we want is a bigger second wave.
Flatten the Curve
We all know we’ve been trying to flatten the curve. In Texas, despite the recent increase in cases, it looks like the curve may be leveling a bit. But remember, this is with extreme control measures in place. New infections in the last two weeks aren’t really shown in these data. Flattening the curve is an ongoing process.
How do we compare to the rest of the country? Bexar county has tested a little bit less of its entire population than the whole of Texas. Texas has tested less than the next most populous states, California and Florida. New York has tested many more, but they started with a much bigger outbreak. The case fatality rate has varied across the states, but that also likely reflects who is being tested. San Antonio has begun testing more asymptomatic people, which will change our denominator.
There are some excellent explanations of Herd Immunity. Is it really a realistic hope? Sure, once we have a vaccine. To achieve herd immunity, at least 60–70% of the population must be immune. The only way we’ve achieved that for other diseases is through vaccination. Even if the current infection rate is three times higher than we see now (assuming we haven’t tested widely), that still means less than 3% infected. With a 3% case fatality rate, that comes to about 294,480 deaths. Those deaths are not collateral damage. Those are people. Our moms, our children, our friends.
Public health isn’t just about medical care. Public health is closely tied to Economic Health. No public health professional worth their degree will say they don’t care if people can’t work. We understand that it is critical for the public to be able to work, to earn a living, to provide food and shelter. This is why it’s so critical that we continue to maintain control measures as much as possible. Going “back to normal” now, without extra infrastructure in place, can flip that downward trend in a flash. And then we’re back to where we were in March. We can re-open in phases, slowly, maintain strict physical distance measures, wearing masks, limiting gatherings while we also invest in public health infrastructure to test, trace, and track new infections. We have no hope of getting to zero any time soon. Our goal is to get to a low level of infection that we can control. We can only do that if the entire public does their part.
*I am assuming the Texas and US data are based on date of report, as it’s not indicated otherwise.