MONDALE'S LAW

 "If you are sure you understand everything that is going on, you are hopelessly confused."- Walter Mondale, former Vice-President of the USA, who died last week.

Just What Is Your Risk of Becoming Seriously Ill Or Dying From COVID, Even After You've Been Fully Vaccinated?

April 19, 2021- by Tom Gill

Reports started to come out last week, both from the CDC and in the news media, of “breakthrough infections,” referring to persons testing positive for COVID, or even being hospitalized or dying from the new coronavirus, even after having been fully vaccinated.  For example:

https://www.cdc.gov/vaccines/covid-19/health-departments/breakthrough-cases.html

https://www.cnn.com/2021/04/14/health/breakthrough-infections-covid-vaccines-cdc/index.html

This may sound alarming at first: and, it goes to a question I’ve had for some time now, “just what is your risk of becoming seriously ill or dying from COVID even after you’ve been fully vaccinated?” The CDC report finally provided some data, and I ran the numbers to try to estimate my answer to that question.  The result?

Becoming fully vaccinated makes car crashes and the flu each at least ten times more dangerous than COVID.

You can stop here- or, if you’d like to see how I delved into the details to get to those numbers, and the fine print, keep reading.

-----------------------------------------------------------------------------------------------------------------------------

From reliable data I have seen, at the time that 66 million persons in the USA had been fully vaccinated against COVID, there were 5,814 reports of “breakthrough infections”- people who got COVID anyway.  However, the vast majority of those cases were asymptomatic, and I’m going to not consider them further because, well, they were asymptomatic (and, besides, there were probably far more asymptomatic cases of COVID-19 in vaccinated persons who hadn’t gotten tested for whatever reason-  being asymptomatic, of course, they generally didn’t have a reason to be tested or know they were COVID-positive!).  Instead, I’m going to focus on people becoming seriously ill to the point of needing hospitalization, or dying, which are pretty standard benchmarks to allow apples-to-apples comparisons to other risks.  396 vaccinated persons were hospitalized, and 74 died: but, 133 of the hospitalized patients were hospitalized for symptoms not related to COVID and found to be positive while in the hospital, and 9 of the fatalities died of something unrelatable to COVID, so I’m discounting those.  That leaves 263 hospitalizations and 65 deaths out of 66 million persons vaccinated.  Thus, at this point (see “Fine Print”),:

·        Being fully vaccinated makes your chances of being hospitalized from COVID to 1 in 251,000.

·        Being fully vaccinated makes your chances of dying from COVID to 1 in a million (actually, one in a million and fifteen thousand).

How does this compare to your chances of being hospitalized or dying from other activities- let’s say, the flu, or from a motor vehicle crash (either as a driver, passenger, or pedestrian, because that’s how the road safety statistics are calculated)?

Let’s look first at comparing your risk to that of a vehicular crash.  The National Highway Traffic Safety Administration statistics for 2018, the most recent data I saw, were 118 fatalities per million persons per year in the USA.  Assuming (and making probably a higher-risk assumption) that the unfortunately afflicted vaccinated people were hospitalized or died one month after being fully immunized,  there would have been 649 vehicular fatalities within that time frame in that group of 66 million vaccinated persons (versus 65 COVID deaths).  Thus, at this point (see “Fine Print”):

·        After being vaccinated, you’re at least ten times more likely to die in a car crash than from COVID.

What about comparing hospitalizations for COVID-after-vaccination to car wrecks?   In the USA, you’re A LOT more likely to be hospitalized, rather than to die, as a result of a motor vehicle incident.  According to my interpretation of CDC data, a person in the United States has a 1 in 1,420 (704 per million) chance of being admitted to the hospital due to a motor vehicle related injury each year.  Again, assuming over a one month period, out of those 66 million vaccinated persons, 3874 persons would have been hospitalized due to car crashes.   Thus, at this point (see “Fine Print”):

·        After being vaccinated, you’re at least fifteen times more likely to be hospitalized from a car crash than from COVID.

Thus,

·        After being vaccinated, motor vehicles are at least ten to fifteen times more dangerous than COVID.

Now, let’s compare the risk of post-vaccination COVID to that of influenza, in a “typical” flu year.  Keep in mind that flu numbers are quite variable: in the months immediately before COVID became a pandemic in 2019-20, the influenza season was quite severe, while in 2020-21, the flu season was extraordinarily mild.  I’m going to use an average to high value for influenza, 49 hospitalizations per 100,000 persons per year in the USA, and as with the crash data, scale it for one month, assuming that on average the COVID patients experienced breakthrough infections one month after being immunized.  Applied to 66 million COVID-vaccinated persons, that results in 2695 persons who would have been expected to have been hospitalized from influenza while 263 were hospitalized with COVID. For influenza mortality, I estimated 11.78 influenza-related deaths per 100,000 persons in the USA per year, scaled to one month as per the other comparisons. This would result in 648 persons dying from the flu while 65 vaccinated persons died from COVID. 

Thus, at this point (see “Fine Print”):

·        After being vaccinated, you’re ten times more likely to be hospitalized or die from the flu than with COVID.  And, thus,

·        After being vaccinated, influenza is ten times more dangerous than COVID.

(So, interestingly enough, car crashes and the flu are about an equal morbidity and mortality risk these days in the USA!): And, all in all, these days, in the USA,

·        After being fully vaccinated, motor vehicles or the flu become about ten times more hazardous to you than COVID.

There are also hints (although definitely not certain) that there are few if any clear cases of persons with (post-vaccination) breakthrough infections that passed the coronavirus on to other persons- so, even if you do get COVID after being vaccinated, it’s probably a dead end for transmission, and you are apparently not as infectious to others.

How do these odds compare to proverbial risks, and where would the current post-vaccination COVID mortality rate rank in regards to other diseases?  Based on my calculations, at this point,

·        The bad news: You’re still more than fourteen times more likely to die from COVID after being vaccinated than to be struck by lightning, and more than 58 times more likely to become seriously ill with COVID after vaccination than to be struck by lightning.

·        The chance of death in the USA from COVID after being vaccinated is roughly the same as the chance of accidentally drowning, or dying from gallbladder disease.

·        After being vaccinated, you’re about five times as likely to be a homicide victim than to die from COVID, and about ten times more likely to die from a gunshot (for any reason) than from COVID.

·        After being vaccinated, COVID becomes about the 55^th leading cause of death in the USA, as opposed to the 3^rd leading cause of death before vaccination.

·        Your chances of being hospitalized for COVID after vaccination are allegedly about the same as your chances of being hospitalized for a fireworks injury, and lower than being admitted to the hospital for unstoppable hiccups or carbon monoxide poisoning.

The rate and numbers of hospital admission for COVID after vaccination are so low that I can’t easily find the data to see where it rates compared to other causes.

-        April 19, 2021: by Thomas E. Gill.

 

 

THE FINE PRINT

These risks and my calculations are extremely fluid and “fuzzy,” and the numbers are “moving targets” and “rough estimates,” for a lot of reasons.

·        These are quick, basic calculations, having to make many assumptions with multiple uncertainties, and given the fact that some of the actual numbers are unclear or unavailable.

·        The COVID breakthrough infection odds for the average American or population as a whole are almost certainly much better than I assumed, because those first 66 million persons vaccinated against COVID included a larger proportion of health care workers (whose jobs put them at greater risk of being exposed to infected people), as well as senior citizens and persons with other illnesses (who will be at greater risk of hospitalization or death if they catch the coronavirus).

·        It’s possible that each individual person, as a result of their unique immune system, may receive slightly more or less immunity to COVID from the vaccination(s) they receive- just like each of us may have a different immune response to being exposed to the SARS-CoV2 virus itself. The jabs may be more or less effective in different groups of people and individuals.  Likewise, individuals will vary in their susceptibility to influenza, vaccinated or not.

·        The risk of the flu is variable depending on which strain is circulating each year, whether or not you and those around you are vaccinated, and how effective the influenza vaccine is from year to year.  Even in the best years, the influenza vaccine is not as effective as the COVID-19 vaccines are now.  The motor vehicle fatality rate is more static from year to year- and there’s no vaccine against being hit by a car or truck. 

·        Your personal risk is modulated quite a bit by your own behaviour- even if vaccinated, are you being careful to not put yourself at greater risk of catching COVID or the flu anyway?  Are you a stay-at-home person mostly in a “bubble” or do you socialize a lot?  Do you work in an isolated office or interact with hundreds of customers?  Do you wear a mask during flu season? Would you have been working at home a lot even if the COVID pandemic didn’t happen, or during flu season?  Do you drive by yourself, ride in a crowded bus,  or walk down streets with busy traffic or isolated roads- and do you do so a lot, or a little?  Are you a defensive driver?  If you’re a passenger, do you wear your seat belts?  If you are a pedestrian, do you keep your eyes on traffic or do you jaywalk?   Even if all of that were equalized,  not matter your own personal behaviour, different cities and states have different rates of flu incidence and traffic injuries and deaths.

·        As more people get vaccinated against the novel coronavirus, the rate of transmission of the disease will decrease and COVID-19 will be occurring less extensively, tending to reduce your risk overall with time.

·        As time goes by, more variants of COVID will emerge, and some of them may be more effective at evading the vaccine, increasing the risk overall with time.

·        After a certain amount of time goes by, the effectiveness of the vaccine will start to diminish, increasing the risk of breakthrough infections.  How long will the mRNA vaccines be effective for?  Scientific studies show it is clearly at least six months: but immunologists and epidemiologists I’ve been following think it’s more likely to be years, maybe even many years.

·        Even things such as flu incidence and automobile-related fatalities were to some extent dependent on the COVID pandemic: as a result of lockdowns, economic slowdowns, masking and social distancing, the most recent flu season was almost nonexistent, and motor vehicle fatalities were down in most areas due to fewer people on the road commuting for much of the year (although I’ve seen statistics saying the rate of vehicular fatalities per mile traveled was way up, because people were more likely to speed and drive recklessly on the emptier roads).   When the pandemic eases, the economy opens up and people start going out, socializing, crowding together, commuting and traveling more, rates of transmissible diseases and motor vehicle injuries and fatalities may increase.

·        A lot more persons are sent to a hospital due to car crashes every year than admitted to a hospital: most are treated and released, and only 9.8% are admitted, according to data I accessed from the National Highway Traffic Safety Administration, Agency for Healthcare Research and Quality, and the CDC.  Since the COVID breakthrough data were for hospital admissions, I used admissions for vehicular injuries as well.

·        Simonsen et al., 2000 (J Infect Dis. 181:831-837) estimated 49 hospitalizations per 100,000 persons for influenza in the USA per year, while Thompson et al. 2004 (JAMA 292(11):1333-1340) estimated 37 hospitalizations per 100,000 per year: these studies were based on different models and different time periods.  The CDC estimated 69 hospitalizations per 100,000 per year for the 2019-20 flu season.

·        For influenza mortality rates in the USA, I used the estimates of Quandelacy et al. 2014 (Amer J Epidemiol 179(2):156-167): other estimates may differ.

·        My estimates of COVID morbidity do not include the potential consequences of “long COVID.” While true, many people are also permanently disabled as a result of motor vehicle injuries, and influenza can also cause long-term side effects.

·        Estimates of the odds of being struck by lightning in the United States are derived from https://www.weather.gov/safety/lightning-odds

·        Death rates for various other causes are derived from: Ahmad and Anderson 2021 (JAMA,  doi:10.1001/jama.2021.5469): CBS News, “Death Index: Top 59 Ways Americans Die,” https://www.cbsnews.com/pictures/death-index-top-50-ways-americans-die/6/ : Murphy et al. 2021 (National Vital Statistics Reports 69(13))

·        The rate and number of hospital admissions for COVID after vaccination is so low that I cannot easily find the data: I could find it in the National Inpatient Sample, but I’d have to pay to access it and might require some extra software.  The claim that 3000 people are hospitalized for fireworks injuries every year (in the USA? It’s not clear) has been made in several press releases findable online.   I found another claim online that 4000 people are hospitalized every year for hiccups (in the USA? It’s not clear).  I have no idea whether it is valid or not.  CDC data states that about 4000 persons are admitted to the hospital every year in the USA for carbon monoxide poisoning.

FRANSEN'S LAW OF FORECASTING

"you'll learn a lot more from your busted forecasts than you will from the ones you hit perfectly (and it's rare to fully hit a forecast perfectly)."

-Tanja Fransen

(Said in regards to weather forecasting- but, really, applies to any kind of forecasting)