COVID-19 Is No Worse Than the Flu — And Other Dangerous Myths
“It’s no worse than the flu.”
“There are lots of respiratory illnesses. People die of them all the time. Why are we making such a big deal about this one?”
These are the kinds of comments I keep hearing from a lot of people. Not just politically motivated comments like, “Trump said it’s a hoax, and he has a hunch based on a couple of conversations that all of that data collected by scientists worldwide is wrong.” No, we’re talking about arguments from data processing professionals and other people who are fact and logic driven. They’re arguing against the #StayAtHome policies and closing schools, churches, restaurants and such on the basis that COVID-19 just isn’t that bad, or that exceptional.
In particular, I had a bit of a knock-down drag-out debate with someone in the healthcare data processing area that I used to work with. He’s pretty sure, among other things, that the data he’s seen in his particular area is the only real data there is. Forget everything from the “lamestream” media. Only he knows the truth.
That got my back up, and I went into full on — here’s the reality, and here’s the data to prove it — mode. After wasting far too much of my life because someone on the internet was wrong, I realized that it wasn’t a completely pointless effort. Put together logically, this tirade against misinformation was a solid informational, myth debunking article.
So, here goes. Let’s start with the big one.
“It’s no worse than the modern flu.”
Not only is that not accurate, it is dangerously misleading. This is the deadliest pandemic since the 1918 influenza pandemic that killed more people at the time than World War I. It’s been 100 years since any single disease has caused this level of death worldwide. An old friend who worked with me in an analytics center of excellence team back in the day, Josh Poduska, the Chief Data Scientist at Domino Data Lab, created a COVID-19 (also known as SARS-CoV-2) tracker using Domino. A lot of data professionals are doing that these days. It helps make sense of the madness, and it’s what we can do to help. Everyone has a strength they can contribute. My strengths are more around explaining data analysis than building it. Hence, this article.
Josh and his current team at Domino, have been doing their best to dispel myths like this using the best weapon against misinformation — real life data. Thanks to the CDC and other health professionals who have been tracking the flu for ages, and Josh’s team compiling that data alongside COVID-19 data, anyone can now see clearly how much deadlier COVID-19 is than the modern flu.
(Updated with more current data and a comparison based specifically on confirmed deaths from each cause according to the CDC. Graph taken from a recent Washington Post article.)
The graph shows deaths in the US over a period of just a few months. And keep in mind that while we currently have a little over 800,000 cases of COVID-19 in the US (Apr 24, 2020), we have about 45 million flu cases in a year. Far more people have gotten the flu in years past, far fewer people died from it.
Just in case that image wasn’t clear enough, the folks at Domo who specialize in data visualization used the high scale analytical database my company Vertica builds to create another COVID-19 tracker with a lot of vivid, clear visualizations that you can click around on to get specific information.
So, no. Not the same as the flu. Far, far deadlier.
There are other differences as well. COVID-19 is more contagious than the flu. There are no vaccines or anti-viral treatments for it. There is very little immunity in the population for it. You can be infected with it for several days, during which you can infect others, before you show any symptoms. In short, it’s a lot more dangerous than the flu.
The sources of these numbers are the US Center for Disease Control (CDC), the World Health Organization (WHO), and Johns Hopkins University, the most reliable sources for disease information in the world.
So, this isn’t any media sensationalism, either left wing or right wing. This isn’t political in any way. This is just the data. Viruses don’t care what your political affiliations are.
You don’t have to take anyone’s word for it. See for yourself.
If you’re into that sort of thing, you can even have a peek at some of Josh’s source code to see how he compiled and presented the data for his tracker.
“These are just ‘reported as COVID-19 cases.’ Anyone can report that someone has COVID-19.”
True, but still extremely misleading. There is a big difference between “reported” and “confirmed.” Confirmed cases are people who were tested, and confirmed to definitely have COVID-19.
Anyone can report a COVID-19 case, a person with all the symptoms, a doctor who spoke over the phone or teleconference with someone who had the symptoms, even a doctor with not enough tests who personally examined someone and believes they have the virus. As long as we don’t have enough tests to go around, the number of actual cases will be higher than the number of confirmed cases.
Most of the trackers I’ve linked to work with confirmed cases. Most importantly, every single one labels the data on the graphs, so if it’s a “reported case,” it will say so. Here’s another outstanding tracker from the Center for Systems Science and Engineering at Johns Hopkins University. It uses Esri to show locality, of cases worldwide as well as in the US county by county. It shows only confirmed cases.
There are two confusing factors here.
One, in the early days of this pandemic, which was horrifyingly only a few months ago, the tests China created were of questionable accuracy. They sold those tests to Europe, so some of the earliest reported confirmed cases might have been misdiagnosed. This means some of the earlier parts of time lines of when this hit and where it hit and how it spread are not fully accurate.
This has already been addressed, and accurate tests are now available. Current numbers of confirmed cases are as accurate as they can be with the second confounding factor.
Confounding factor number two: there aren’t enough tests. I personally know one person in Austin, Texas who had all the symptoms, passed a questionnaire with a nurse, and was told to just stay home and treat symptoms unless it got worse. When she asked why she couldn’t get tested, she was told, the hospital system didn’t have enough tests for everyone. The person arguing with me said they also knew twenty other people who hadn’t been tested when showing the symptoms, for the same reason.
The rapid spread of COVID-19 has overwhelmed our healthcare system, and there just aren’t enough tests to go around. Basic triage says, save the few tests you have for people at higher risk. So, yes, many people are not being tested who should be, and cases are being underreported.
What does this mean? The data isn’t accurate? All those graphs I keep pointing you at are wrong?
Yes, and absolutely no. Since every case you see on those graphs is confirmed, unless it says otherwise, that is reality. That is accurate. Every single person who those graphs say has COVID-19 or died of COVID-19 or recovered from COVID-19 did. That is 100% true, according to our best, current testing capabilities.
But, it could be a lot worse than the graph would make you think. There are a whole heck of a lot more people out there with COVID-19 symptoms who aren’t being tested. This means many people might have it, take care of themselves at home and get over it, and never become confirmed cases. It also means many people might die of it and never become confirmed cases.
It means the numbers of dead that you see, the numbers that are terrifying and unprecedented for a hundred years, are the minimum number of people dead by this virus. The reality is worse, possibly far worse.
It means the incidence of people around you who have the virus and are carrying it may be way higher than the numbers on the charts indicate.
It means that the COVID-19 pandemic could be a lot worse than it looks.
John Dormer is another data professional who jumped into the debate. He has been posting a daily tracker for his friends on what the latest changes worldwide are as far as getting the contagion under control, based on his own analysis. He’s more of a numbers kind of guy than a graph visualization kind of guy, so the results of his analyses are in a spreadsheet. He’s using the Johns Hopkins data on github.
John Dormer — “What’s hell to model is governmental and societal responses. It is possible that the perception that all hospitals are overrun took hold in the population, suppressing trips which might have led to diagnosis. Local governmental and media reports may have modified behavior. People being more cautious might have caused a genuine drop in infections, which led to fewer diagnoses. Without a way to flash-sample a complete population at a moment in time, it’s impossible to know without significant error.”
On the plus side, if way more people have it, and the death numbers are at least close to the same, it means a lower percentage of people who get it will die. The only way we’ll really know the full truth is if there are finally enough tests for everyone. Even folks who have already died should be tested if they weren’t before death, to get an accurate count of just how deadly COVID-19 really is.
But those complicating factors don’t change that this thing is highly contagious and it’s killing people in record numbers.
This doesn’t mean panic, and it doesn’t mean you’re doomed to die if you get it. The survival rate is still very high, even in older patients who are most at risk. I personally know two people at my company HQ in Boston who got it, and got better after a few weeks. It’s not a death sentence. For most, the effects are similar to a bad case of the flu.
But I also know of a 17-year-old track runner who is currently on a ventilator fighting for his life.
It’s not a death sentence, but it is also not even remotely business as usual.
“There appears to be a peak late last year in respiratory illnesses, many that caused death, yet tested negative for the flu. No one made a big deal out of those. So, we shouldn’t make a big deal out of this.”
This has some validity. It’s possible we’re actually on a second, more serious wave of this virus, rather than on the first wave as we thought. Of course, we won’t know for sure until that localized data is correlated with other data sets. But for tracking the virus’ origins and spread, this is important information. I would be interested to know if it’s verified by data from other locations. No conclusion can be made yet, but it’s worth digging into. It would certainly mess up the folks who are blaming this whole thing on China because they happened to be the first place an outbreak was seen and tracked.
On the other hand, it has zero impact on how we should handle the current crisis.
“None of this is a valid argument to let up on modifying our public behaviors. The only model we have for this scale is the 1918 influenza pandemic, and it indicates that human nature is unchanged in 100 years. We’re in this for the long run. With the relaxation of isolation protocols, we will see a sharp increase in infections, a respectable increase in detections, and an intolerable one in deaths. We don’t know if immunity afterward is robust, or only robust in certain people. We don’t know if there’s any long-term relapse, like with chicken pox and shingles.” — John Dormer
We’re seeing a leveling off of new cases, a slight drop even. The data says what we’re doing is working. The only way to tell for sure would be to have more tests, but meanwhile, let’s keep doing the one thing that looks like it’s keeping more people from dying.
“People die all the time. New York City, for instance, it’s no different now than any other time.”
I was the one that brought up New York City. Their death toll is the worst of any single city in the US. There are whole countries that haven’t lost as many. 16,388 confirmed dead from COVID-19 as of April 24, 2020 just in the city.
Ironically, New York is one of the states with the highest rate of testing, after it was posited as an argument before that the low level of testing somehow means confirmed positive results are meaningless. The premise being that people confirmed to positively be sick or dying or dead from this virus don’t count because everyone wasn’t tested for it. Incomplete data doesn’t count. This would mean that pretty much every study ever done was useless, and we should generally just not use data and go back to following hunches. When data is incomplete, regardless of the reason, we note that, and alter our conclusions and possible error levels based on the data we don’t have and why we don’t have it. We don’t just toss out the existing data.
That said, New York’s data is closer to complete than any state, other than Rhode Island.
The states that test the most, show the highest rates of infection. That’s kind of a chicken egg statement, because maybe they test more because they have more cases (and access to more tests), or maybe they show more confirmed cases because they test more.
As of April 24,2020, the top 4 states for testing are:
Rhode Island, New York, Louisiana, Massachusetts.
New York has had 695,920 people tested, 276, 711 confirmed cases, 21,283 deaths in the whole state. That’s insane and terrifying, and as I said before, the biggest chunk of deaths in any city in the US is in New York City. They’re dealing with a high rate of hospitalizations, a staggering death toll and an overburdened healthcare system.
But that’s like a normal day in NYC, right?
Oh, hell no. Not even close.
In a normal day in NYC, about 145 people die. In April of 2017 and 2018, 146–149 folks died in New York City every day. On a bad day, 159 die. That’s from every cause from a 97-yr-old with a heart attack to a 25-year-old marathon runner who stepped out in front of a bus. 145 people dying of all causes is normal for New York City.
“The first five days of April saw 1000 people die of the coronavirus alone in New York City. That’s more than 208 per day” from one cause alone — COVID-19.
New York is setting up mobile morgues. Funeral homes are overwhelmed. (Data source — Summary of Vital Statistics 2017, Data Source — NYC health Covid-19 data)
“But what about this one study that says lots of people have it and are fine?”
“A few days on, and the comments on the article are quite properly trouncing this study. The rate of expected false positives, once correctly calculated and applied, covers most of the observed positive tests, rendering the study meaningless. The point that I agree on is that reporting is neither being handled correctly nor is testing sufficient in scale. This is leading people to incorrect conclusions.” — John Dormer
“Am I supposed to just believe everything CNN tells me?”
This age of massive distrust of news, not of just one source of news, but all sources of news, blows my mind. Reporters can go off the rails trying to convince you of something, they can see things from their own point of view, and they can often tend to focus on the most unusual and interesting parts of the world, which can lead you to believe that those are all there is, but they are not out to lie to you.
That’s not what the overwhelming majority of reporters studied journalism for.
People who become reporters, in general, have a powerful need to tell the truth. And not just tell the truth, but shout it from the rooftops, walk into enemy fire for it, stand next to burning buildings for it, accept abuse and assault for it, get thrown into jail for it, stand out in hurricane winds for it. Truth is a reporter’s religion. Their entire job and purpose in our society is to dig up truth, even unpleasant truth, or truth that is hidden, and give it to you.
If you suspect a specific news source is flat wrong, or giving you misleading information on purpose, then check three or four other reputable sources for verification. If they all say the same thing, then YES, believe them already.
Truth doesn’t stop being truth because you don’t like the guy saying it.
That said, screw the news media. You don’t have to just believe what they tell you in this case. Look at the data for yourself from the same reputable scientific sources the reporters are using: WHO, CDC, Johns Hopkins, ECDC, NHC, DXY, the COVID Tracking Project, and so on. They’re all saying the same thing. Believe the evidence of your own eyes.
Truth doesn’t stop being truth because you don’t like it.
Protect the people you care about. If you get infected with this virus, chances are high that you’ll spread it to the people closest to you who are most vulnerable. So, stay safe. You’re not just protecting yourself, you’re protecting everyone around you. Use precautions. Don’t let anyone tell you this is no worse than the flu. They’re lying, and you can see that now for yourself by looking at the data.
Do what you need to do to not become a statistic.
