The risk from Covid-19: killer #3, after #1 heart disease and #2 cancer.
Almost everyone knows someone who died from heart attack or cancer. In 2020, more than half of Americans will know someone who died from Covid-19. But fewer than half will know for sure that the person died from the disease.
The current (as of 8/9/2020) U.S. Covid-19 mortality rate, according to Johns Hopkins, is ~50 (per 100,000.) This is an underestimate. The actual rate is over 64. Because this figure is analyzed on an annual basis, the real rate will be at least 100. Current Covid-19 deaths are perhaps 2/3 of the total for 2020.
This isn’t about politics. The evidence is clear. I’ll provide it below.
Each of us has, on average, a friend and acquaintance network of about 600 people. This includes friends from the past, extended family, people you don’t see much but have known. 60% of us will know someone in our extended network who died from Covid-19, by the end of the year. It probably won’t be close friend or family, but someone we knew from a previous job, or a neighbor we see occasionally.
Acquaintance networks grow with age, and peak before retirement, and Covid-19 kills more people as they get older. So older people are more likely to know someone who died than younger people.
Because flu pandemics are the closest similar phenomena to Covid-19, how does flu mortality compare? Flu deaths may be recorded as pneumonia or other complications. The CDC reviewed evidence and decided to include a percent of other respiratory deaths as flu-caused. Doshi wrote a critical paper. He estimated monthly figures of “just influenza” mortality rates (per 100,000.)1 They rose as high as 23.3 in Jan. 1969, but his rates are usually in low single digits, every year.
The consensus, however, is that “just flu” figures are inaccurate. The CDC’s data reflects the consensus.2 It estimates that in the 2017–2018 flu season, the mortality rate was 19.1 (although it only provides rates for different age groups on its website.) The over 65 flu mortality rate was 100.1 that year. These numbers are for the four or five months that define the flu season. That is, about 62,000 people died from flu and flu-caused complications in the four or five months of the 2017–2018 flu season.
The same under-counting happens with Covid-19, as used to happen with influenza. Deaths from pneumonia and other Covid-19 caused complications may not be reported as Covid-19 deaths, nor are deaths that occur in non-healthcare environments.
The CDC estimates that excess deaths attributable to Covid-19 are about 25% to 30% more than those actually recorded as Covid-19. As of 8/9/2020, 162,423 Americans have officially died from it. Add 28%, the total is about 208,000 deaths. Divide the US population of 327,000,000 by 208,000, and the current mortality figure is 64 per 100,000.
Because most of the news media reports do not reflect this total, I recognize that some will doubt it. I invite readers to review the CDC link. Further, if we are to compare Covid-19 mortality with influenza, and the flu rates include both reported and unreported deaths, then we have to apply the same metric to Covid-19.
As explained, the Covid-19 season continues, and mortality increases. The number of infected could double by year’s end, but treatment of those infected is steadily improving. Conservatively, the combined reported and unreported mortality rate will reach 100 per 100,000. Five times the 2017–2018 flu season, which was the worst in the past decade.
I invite anyone to review the data above, and explain how my figures are off. This is not a political or ideological review. It’s an attempt to determine the baseline threat from Covid-19, to better estimate risk. If mortality in the general population is .001, and each of us knows 600 people, then each knows .6 persons who died, on average.
Comparing Covid-19 to the flu is imperfect. Most people don’t know anyone who died from the flu. Instead, I consider auto deaths. According to the National Safety Council, auto-related mortality rate per 100,000 Americans is 12. Covid-19 is eight times more dangerous than riding in a car.
The auto death rate was nearly 50 in the late 1960s, according to the FHWA. People didn’t stop driving in the 1960s, but Ralph Nader’s Unsafe At Any Speed was published in 1965. It led to seat-belt laws, the creation of the US Dept. of Transportation, and the precursor of the National Highway Traffic Safety Administration. Nader’s crusade resonated with Americans because, in fact, a rate of 45 or 48 deaths per 100,000 meant people did experience the trauma of knowing someone who died in a car accident.
The mean social network size (how many people is someone acquainted with, on average) is about 600. An annual death rate of 50 means 3 out of 10 people, or roughly every third person, knows someone who died, each year, in a car crash. Over a multi-year period, the majority of people do.
A report similar to Nader’s was published in the late 1950s, but made much less of an impact. Although Nader’s writing style and charisma may have played a role, the car accident mortality rate in the late 1950s was much lower, in the mid-20s. Actually lower than flu mortality in some years. On average people were half as likely to know someone who died, compared to a decade later.
Covid-19’s mortality rate is at least twice what auto fatality rates were when they peaked 50 years ago. Very roughly, the majority of Americans will have one acquaintance who dies from it. Of course, this varies by age group, ethnicity, location, and other factors.
People didn’t get into a car in 1968 and expect to die. By the time seat belt laws were passed, they really didn’t expect to. There was significant opposition to seat belt laws. 65% opposed them in 1984. By then, the car mortality rate had fallen to 19.4, similar to a bad flu year. Most people no longer had contact with anyone who died in a car accident.
Today about 20% oppose wearing facemasks. They’re visible, because those same people are likely to go out and mix. But 20% is 1/3 of the opposition to seat-belts, which isn’t surprising, since Covid-19 mortality is 5 times what the car mortality rate was in 1984.
People sense the greater risk from Covid-19.
Is there anything Americans experience that feels more dangerous than Covid-19?
About 1/4 of traffic fatalities involve someone drunk. The total fraction of drivers who are drunk, at any one time, is much less than 1/4. About 134 million Americans drive each weekday. Roughly 21.4 billion days of driving occur in the US, annually. Gallup, extrapolating from an anonymous survey asking Americans to admit how frequently they drive drunk, estimated about 111 million days of drunk driving occur in the US annually.
This seems an underestimate. At that rate, 1 out of 200 drivers would be causing 1 out of 40,000 people to die, 25% of annual roadway fatalities.
If we double the drunk driving total, 1% of drivers are drunk. That means 1,000 of every 100,000 drivers cause 25 of every 1,000 people to die. If these are all drunk drivers dying, their death rate is 2.5%. 2,500 of every 100,000.
Most people don’t know many drunk drivers. Still, driving drunk is five times more dangerous than Covid-19. If you think about it, that’s pretty serious. Most of us don’t drive drunk, and would be afraid of being a passenger in a car where someone was drunk. But Covid-19 mortality varies by age, and people age 70 are 15 times more likely to die from the disease than the average person. At 70, Covid-19 risk is three times greater than driving drunk.
The media and politics has, in the past, exaggerated risks from terrorism, so that too many believe it’s a serious threat to themselves and their loved ones. (In fact, the chance of terrorist caused death in the US, including 9/11, is under 1 per 100,000.) On the other hand, other risks, such as from coal-fired power plants, are underappreciated.3 In both these cases, right-wing politicians pushed or suppressed risk. So it’s not surprising to see right-wing leaders again try to manipulate risk.
To say that Covid-19 is about 8 times more deadly than driving a car isn’t that useful, since Covid’s mortality rate varies by age much more than driving fatalities do. The fatality rate of someone 40, which is close to the mean death rate for the entire population, goes up 15 times by age 70. Someone in their 20s is 1/4 as likely as someone 50 to die. Their Covid-19 mortality risk is only a little higher than their auto death rate.
Of course you can’t transmit a car crash injury, and age isn’t the only related variable. But this helps explain the attitudes of many younger people.
Here’s a list of 2017 mortality, by cause, in the US:
Heart disease: 647,457
Cancer: 599,108
Covid-19 2020 estimate: 300,000
Accidents (unintentional injuries): 169,936
Chronic lower respiratory diseases: 160,201
Stroke (cerebrovascular diseases): 146,383
Alzheimer’s disease: 121,404
Diabetes: 83,564
Influenza and Pneumonia: 55,672
Nephritis, nephrotic syndrome and nephrosis: 50,633
Intentional self-harm (suicide): 47,173
Covid-19 US mortality could reach 300,000 by the end of 2020. Like heart disease and cancer, Covid-19 mostly kills older people. Over half of the US population will have an acquaintance who dies from it, but perhaps only 40% will know for sure that they died from Covid-19, because their death may be attributed to something else.
1 Doshi, Peter. “Trends in recorded influenza mortality: United States, 1900–2004.” American journal of public health vol. 98,5 (2008): 939–45. doi:10.2105/AJPH.2007.119933
2 Thompson, William W et al. “Estimating influenza-associated deaths in the United States.” American journal of public health vol. 99 Suppl 2,Suppl 2 (2009): S225–30. doi:10.2105/AJPH.2008.151944
3 Correia, Andrew W.a; Pope, C. Arden IIIb; Dockery, Douglas W.c; Wang, Yuna; Ezzati, Majidd; Dominici, Francescaa Effect of Air Pollution Control on Life Expectancy in the United States: An Analysis of 545 U.S. Counties for the Period from 2000 to 2007, Epidemiology: January 2013 — Volume 24 — Issue 1 — p 23–31 doi: 10.1097/EDE.0b013e3182770237
