All right, it’s 225 deaths. What’s the big deal?
I’m fully aware the skeptics of COVID-19 data skepticism have looked at the headline, looked at the death toll in the United States — 156,806 as of Wednesday morning — and are wondering why a difference of 225 deaths in Texas should even matter.
So let’s state right off that, while every death is a tragedy, in the scale of the coronavirus disaster, the number we’re about to talk about is a drop in the bucket. It’s infinitesimal. In isolation, it changes nothing. Even outside of isolation, it doesn’t change how seriously we should take the novel coronavirus.
To understand the macro, however, we need to look at the micro.
On July 27, the Texas Department of State Health Services issued a correction on the state’s death toll, revising it downward by 225 deaths.
The move came after the DSHS implemented a change in reporting COVID-19 deaths, which came as the state has been hit hard by a second wave of the virus.
“The Texas Department of State Health Services is improving the reporting of fatalities due to COVID-19 by identifying them through the cause of death listed on death certificates,” a news release stated.
“This method allows fatalities to be counted faster with more comprehensive demographic data. Using death certificates also ensures consistent reporting across the state and enables DSHS to display fatalities by date of death, providing the public with more information about when deaths occurred.”
The count had previously relied on “local and regional health departments after they received a notification and verified the death.” However, given that different jurisdictions take different amounts of time to report the cause of death, the department decided to make the process more timely.
Now, “[a] fatality is counted as due to COVID-19 when the medical certifier, usually a doctor with direct knowledge of the patient, determines COVID-19 directly caused the death. This method does not include deaths of people who had COVID-19 but died of an unrelated cause. Death certificates are required by law to be filed within 10 days.”
Under the new method, the number of those who had died of COVID-19 was revised from 5,038 to 5,713, a jump of 675 deaths.
But that wasn’t right, either. It turns out an automation error led to 225 more deaths being added to the new total than should have been, as the DSHS announced on July 30:
DSHS corrects COVID-19 fatality counts for the week of July 27.
An automation error caused 225 fatalities to be included even though COVID-19 was not listed as a direct cause of death on the death certificate. #COVID19TX dashboard: https://t.co/ofycOLqWQZ pic.twitter.com/4mKBzjIrfO
— Texas DSHS (@TexasDSHS) July 30, 2020
The left-leaning digital publication The Texas Tribune lamented the change was “lending fuel to skeptics who question the accuracy of the government data.” However, the Tribune’s article simply notes what most of the responsible “skeptics” have been saying all along: That data collection so far is limited and error-prone.
This is understandable, to a certain extent, regarding a new and deadly disease like COVID-19. The problem is that we have politicians making life-and-death decisions based on that data.
Less than two weeks before this error, Texas was forced to remove 3,000 cases from its positive case count due to testing in San Antonio:
— Ted Cruz (@tedcruz) July 17, 2020
The reason, according to the Austin Statesman, is that these were actually cases counted via antigen tests. According to U.S. News, antigen tests produce fast results by detecting fragments of proteins from the virus that are present in nasal swabs of test subjects.
Another type of test — known as a polymerase chain reaction test or PCR — detects the presence of coronavirus genetic material in a nasal swab.
“Antigen tests are very specific for the virus, but are not as sensitive as molecular PCR tests,” stated a May memo on the tests, according to the Statesman.
“This means that positive results from antigen tests are highly accurate, but there is a higher chance of false negatives, so negative results do not rule out infection.”
Under Texas’ standards, an antigen test must be followed up by a molecular PCR test to be counted in the confirmed case category.
“The case data on our website reflect confirmed cases, and cases identified by antigen testing are considered probable cases under the national case definition,” a DSHS spokesman said.
Thus, these cases were removed, but the fact that the antigen test results in false negatives rather than false positives likely means there were more cases in San Antonio than were reported, not fewer.
Meanwhile, The Associated Press reported on July 21 that Connecticut’s public health laboratory reported 90 false positives from one manufacturer’s test due to a flaw. That number doesn’t sound like a lot, but the test is used throughout the country.
And then there was Florida, which — being Florida — managed to make headlines when it reported deaths from COVID-19 where the decedents had passed due to, among other things, a motorcycle accident and a gunshot to the head.
Writing in March, as the scope of the pandemic became clear, noted Stanford academic Dr. John Ioannidis declared that COVID-19 may “be a once-in-a-century evidence fiasco.”
“At a time when everyone needs better information, from disease modelers and governments to people quarantined or just social distancing, we lack reliable evidence on how many people have been infected with SARS-CoV-2 or who continue to become infected,” Ioannidis wrote in STAT, a health-oriented website. “Better information is needed to guide decisions and actions of monumental significance and to monitor their impact.”
Several paragraphs later he stated the point plainly:
“The data collected so far on how many people are infected and how the epidemic is evolving are utterly unreliable. Given the limited testing to date, some deaths and probably the vast majority of infections due to SARS-CoV-2 are being missed. We don’t know if we are failing to capture infections by a factor of three or 300.”
Despite countless stories about the unreliability of coronavirus infection reporting, the general stated opinion among experts is that we’re undercounting both COVID-19 cases and deaths, as the Texas Tribune noted. By how much? That answer varies widely, because even they don’t know. And they admit the data are unreliable. We’re supposed to trust them, though — and the idea that the data are improving rapidly.
The data may be improving, but the question remains how we’re counting deaths by COVID-19.
In a July interview with the Greek Reporter, Ioannidis noted “that the vast majority of people who die with a COVID-19 label have at least one and typically many other comorbidities. This means that often they have other reasons that would lead them to death. The relative contribution of COVID-19 needs very careful audit and evaluation of medical records.”
“The number of COVID-19 deaths can be both undercounted and overcounted, and the relative ratio of over- and under-counting varies across different locations,” Ioannidis told the Greek Reporter. “In most European countries and the USA it is more likely to be overcounted, especially if we are talking about ‘deaths by COVID-19.'”
Ioannidis may be an outlier in this debate, but this is part of the problem: To what extent should deaths be attributed to COVID-19 when the decedents have significant comorbidities? And to what extent are decisions that are being made actually making other problems worse?
In a scathing article in the May/June issue of the Hillsdale College publication Imprimis, conservative commentator and scholar Heather Mac Donald notes that the author of the Imperial College model — the most dire of the major coronavirus models — has acknowledged that, under his model, two-thirds of COVID-19 victims would have died of their comorbidities by the end of 2020.
The headline on MacDonald’s piece was: “Four Months of Unprecedented Government Malfeasance.”
In the United States, over 40 percent of deaths have been in nursing homes, according to The New York Times. While this has certainly been exacerbated by bad policy, it’s also fair to say that’s where you would find the most comorbidities.
This is quite the rabbit hole over an “automation error” that miscounted 225 deaths in the face of a global tragedy, but that’s the point. I’m not bringing in extraneous cases to intentionally muddy the water and intentionally confuse the issue. I’m merely pointing out how muddy the water really is.
In the absence of a vaccine, we clamor for more testing and better governance. Those are important, but we also ought to be demanding better, more consistent data.
An “automation error” in Texas wouldn’t be such a big deal if those 225 deaths weren’t part of a much wider problem.
Truth and Accuracy
We are committed to truth and accuracy in all of our journalism. Read our editorial standards.