The Coronavirus Fatality Totals Aren’t What They Seem.
Here’s some troubling information (since we all could use a little more of that right now). The news stories displaying an ever-increasing tally of the “death toll” from Covid-19 only give a small sliver of the actual fatality figures. These in-the-moment numbers are limited to the “laboratory-confirmed death” reports. By analogy, imagine the death toll a soldier in a war zone can ascertain from a foxhole while the bombs are still dropping. Only when the bullets stop flying can the true damage assessment occur.
I’m not bringing this up to be needlessly depressing. Unlike any other national crisis in recent memory, our country’s response hinges on millions of people drastically modifying their daily behavior — and that won’t happen if people are confused about the actual gravity of the threat. And a major source of confusion right now stems from faulty comparisons between our early Covid-19 fatality data and the statistically projected fatality data we have from past outbreaks.
Here’s what I mean:
The fatality numbers from past outbreaks (e.g. flu, swine flu) that many people are using to gauge the seriousness of the coronavirus outbreak are retrospective statistical projections researchers made once the dust had settled and more complete data was available — often resulting in statistical projections revealing 10–15 times as many deaths (or more) as the lab-confirmed cases that were observed in real time.
Swine Flu (H1N1)
For example, when you hear, “The coronavirus has killed 1,200 Americans, but 17,000 Americans died from the swine flu in 2009–2010, so why all the panic?” what you aren’t hearing is what the in-the-moment, lab-confirmed death count was while the swine flu was spreading — the equivalent of the daily “death toll” the news is currently tracking for Covid-19.
Here’s what that real-time fatality count was for the swine flu during the middle of that outbreak: During the second wave of the swine flu (Aug. 2009 to April 2010), the number of lab-confirmed fatalities was just 2,096. For context, the US is on track to exceed the 2,096 fatality figure for lab-confirmed coronavirus deaths in just the next few days. Only with the benefit of time and research were experts able to project the actual H1N1 death toll. Instead of 2,000+/- fatalities during the second H1N1 wave, experts have projected the actual figure at around 14,800 fatalities. Roughly seven times more than the lab-confirmed count.
But going back in time a decade to the swine flu outbreak, without the benefit of retrospective analysis, the news media reported the purported death toll based on the limited data that was available. By the fall of 2009, some news reports pegged the H1N1 death toll as something “more than 1,000” — a far cry from the tens of thousands that would ultimately be projected. Similarly, the CDC reported only a few hundred lab-confirmed fatalities during the outbreak, while emphasizing that the lab-confirmed cases “represent an undercount.” Only as months of the outbreak wore on were the more accurate projections cited today created.
The Flu (Influenza)
We see a similar variance between the lab-confirmed cases and the statistically projected cases with the common flu. Just as with H1N1, saying, “37,000 people died of the flu last year, and only 1,200 Americans have died of coronavirus,” reflects the same faulty data comparison. For the 2018–19 flu season, the CDC reported only 19,543 laboratory-confirmed influenza-associated hospitalizations — but projected the actual number to be around 20 times that high, at 400,000. Similarly, only 155 lab-confirmed pediatric flu deaths were reported, while the actual projected figure was far higher.
The Coronavirus (Covid-19)
We likely won’t have anything resembling an accurate projection of the true death toll of the Covid-19 pandemic for months, if not years. In the meantime, we need to be mindful of how we contextualize the premature fatality data that we have in order to avoid conveying a false sense of security.
Listen to the scientists. Stay at home.
