Do Not Compare COVID-19 (Coronavirus) To Seasonal Influenza

[42pines]

Covid-19 should not be compared to seasonal Influenza

It’s normal to look at the CFR (case fatality rate) of seasonal influenza when considering Covid-19. After all, 20-45 million get “the flu” every year. The numbers vary depending on the year. The “low” was the 2011-2012 season with 9.3 million cases and 12,000 deaths and the high was the 2017-2018 season with 45 million cases and 61,000 deaths._¹

Let us look at the worst of eight recent flu seasons. The CFR for that was 0.0014 meaning, 1.4 persons died per 1000 cases. Interestingly, if we look at the “best” of the past eight seasons (the “good” one) the CFR is 0.0013, almost the same number of deaths per thousand cases.

So, hold on to this thought...

Seasonal influenza’s average risk of dying is about one death per 750 cases.

I see and hear CFR for the Covid-19 repeatedly being quoted at around 2-3%. CFR currently (as of 2/27 12:23AM EST) is 3.4%. That is 2,810 deaths divided by 82,550 cases.

But to say that the CFR for Covid-19 is 3.4% is not wise. Why? Simply put, the CFR is not applicable are even a number that one can calculate since it's a "look-back" number.

For data current as of 2/27, click here.

CFR is a “look-back” rate

It is very useful when we look back to an epidemic/pandemic that has ended. For instance, when comparing the “good” 2011-2012 flu season to this season’s numbers. CFR is death rate / # of confirmed cases. To divide the known number of cases today by the number of deaths, we get 3.4%. But this is not wise to use because it assumes that ALL of the remaining cases (total cases today of 82,550, minus cases where the person has recovered or died) will be cured. This is “magical thinking.” Many of those remaining 46,488 will recover, but some will die.

In short, we won’t know the true CFR for another year or more. CFR is useless at this time, it will be helpful sometime next year.

Known outcomes

So, let us look at the cases with known outcomes, meaning the sum of the cases of those who have recovered plus those who have died. This population is composed of those 33,252 who have, to today, recovered plus the 2,810 who have died; thus, our population is 36,062. If we divide 2,810 (deaths) into our population of 33,252, we find an overall death rate of 8.45% or about 85 per one thousand cases with known outcome.

Please let that sink in for a bit. Today, looking at Covid-19 cases with known outcomes, 85 out of 1000 cases have died.

Should this hold, and it may, then 1 out of 12 cases will die.

Compare this to seasonal influenza where 1 out of 750 cases will die.

The death rate currently for Covid-19 is 62x the death rate of the average seasonal influenza.

These two illnesses do not compare, although we’ll really not know for a year or so, at which time we can look back to see what the true CFR rate was.

Let us look now at areas/countries to see the current death rate is, of those cases with known outcomes.

I’ll call the current fatality rate of known outcome cases: CFR-KO.

I shall call the population with known outcomes: PWKO

Total deaths / PWKO = CFR-KO or current death rate.

• World: 2801 / 36,062 = 8.45% or ~85 deaths per 1000 cases.
• Hubei Province: 2641 / 26,024 = 10% or ~100 deaths per 1000 cases.
• South Korea: 13/35 = ~37.2% or 372 deaths per 1000 cases.
• Italy: 14/54 = ~26% or 260 deaths per 1000 cases.

We see that regional epidemic outbreaks tend to drop in CFR-KO as time goes by. Whether the true “look-back” CFR will drop to 2 or 3% or some other number (lesser or higher) depends on many factors which are simply unknown. For instance, children appear to be much less at risk for Covid-19, and cases within Africa (other than the one in Egypt) seem oddly absent. Perhaps the virus does not do well in warmer climates. Perhaps the fatality rate will drop as we approach Summer in the Northern hemisphere—we do not know.

Perhaps the tendency of the media to quote the 2-3% theoretical CFR comes from the fairly recent 2009 H1N1 flu pandemic. The 2009 “swine” flu was very feared. It infected 10-200 million world-wide with estimates of death ranging from 105,700 to 395,600 and a CFR of 0.03% or 3%. (30 per 1,000 cases). (Wikipedia). It was expected to return far, far worse numbers.

The “swine” flu, which was the first pandemic of this century, turned out to be far less lethal than many expected. Triggle wrote in an article from December 10, 2009 titled, Swine flu less lethal than feared, "The swine flu pandemic is ‘considerably less lethal’ than feared, chief medical officer Sir Liam Donaldson says,” adding, “A study led by Sir Liam found a death rate of just 0.026% in those infected, the British Medical Journal reported.”²

Will we be lucky?

Perhaps we’ll be lucky this time, too, although I suspect we won’t be as lucky as we were for the “swine flu.”

[Susie2310]

1 hour ago, Susie2310 said:

42pines, it seems that you have significant education/training in statistical methodology and epidemiology. Your OP says that you specialize in Occupational Health, and I recall your OP also says that you have a degree in Chemistry. Would you be willing to tell us about the education and courses you have taken, especially your education/training in statistics? My understanding is that you are also a nurse. Am I understanding correctly that you are an Occupational Health Nurse (or do you work in Occupational Health in another role) and also work in the ICU as a nurse? How did you decide to enter the field of Occupational Health?

In regard to your OP and to your other comments on this thread, what is your opinion of the data that has been collected/publicized so far and the analyses of the data, including your reasoning/evidence for your opinions?

[Kyrshamarks, BSN, RN]

And people actually believe the numbers coming out from the Chinese government? My wife has a niece (she is from the Philippines) who is teachih in China. From what she has told us, the death toll is way, WAY higher than they are letting on. Think in the upper 5 to 6 figures. Not 2000+..

[myoglobin, ASN, BSN, MSN]

Not to make the case for socialized medicine but the Netherlands, Germany, Norway and Sweden have around double the cases (confirmed) as in the United States, but haven't recorded a single death between them. What are they doing that our King County health system is failing to accomplish (keeping in mind that most of the deaths here are from older people many of whom are from a single institution)?

[Nurse and Spy, BSN]

Thanks for this terrific article. And yikes, today I heard Dr. Peter Hotez say they are finding the fatality rate for older adults is 10-15%. He literally called it an "Angel of Death" for older people.

I also found this quote in an article (https://www.foxnews.com/opinion/Dr-peter-hotez-west-coast-coronavirus-outbreak-highlights-3-vulnerabilities-we-need-to-urgently-address)

He wrote on the topic: "Reports from China in January and February found that the SARS CoV2 disproportionately kills individuals over the age of 60, with those greater than 80 years of age suffering mortality rates as high as 15 percent."

And this: " We urgently need federal guidance from the CDC on how to best design and implement infection control practices for these facilities and prevent the spread of SARS CoV2 – a tragic and scary “angel of death” among our senior citizens."

His credentials: https://www.bcm.edu/people/view/peter-hotez-m-d-ph-d/b1846a47-ffed-11e2-be68-080027880ca6

[myoglobin, ASN, BSN, MSN]

I believe that the fatality rate is likely much lower than the 3% given by the WHO based on the fact that so few people can even be tested given a lack of covid 19 testing. Stated differently there are probably exponentially more people who are or have been infected with covid 19 than has been officially recognized.

[NurseBlaq]

On 3/9/2020 at 8:21 AM, myoglobin said:

I believe that the fatality rate is likely much lower than the 3% given by the WHO based on the fact that so few people can even be tested given a lack of covid 19 testing. Stated differently there are probably exponentially more people who are or have been infected with covid 19 than has been officially recognized.

This has been my line of thinking as well.

[42pines]

From the article: “'Don't believe the numbers you see': Johns Hopkins professor says up to 500,000 Americans have coronavirus” and from the article: 'Don't believe the numbers you see': Johns Hopkins professor says up to 500,000 Americans have coronavirus.”

“Ohio Department of Health Director Amy Acton said at a press conference alongside Gov. Mike DeWine (R) that given that the virus is spreading in the community in Ohio, she estimates at least 1 percent of the population in the state has the virus.” “"We have 11.7 million people. So, the math is over 100,000. So that just gives you a sense of how this virus spreads and is spreading quickly."

I’ve never been more disgusted with our government and our public health system. I wrote to my Senators on January 30th, and never even received an answer. I’m disgusted that there has been little action, it smacks of utter incompetence. One of the Boston hospitals wanted to develop their own test and indicated that they could, in fact do this, but FDC said “no.”

The reality is that it does not matter if the eventually CFR turns out to be 3% or Italy’s 7% or 1% or ½% I really don’t think this will turn out to be “just the flu.”
To all you nurses out there on the front line, (however, there is no front line here, this will touch pretty much all nurses and Practitioners) take care!

[2BS Nurse, BSN]

On 3/2/2020 at 12:13 PM, Nurse SMS said:

Timely and well outlined. I do wonder, though, at the "known cases" portion of the equation. With the limitations of available testing and the likelihood that many are infected and only get a mild case or even no symptoms, these mortality numbers may be skewed high.

I agree. Right now, our "known cases" are going to be the most severe because they were allowed to be tested. The mild patients (basically, if you can talk you don't need to be tested, you need to go home) are not being added into the "known cases" count.

[2BS Nurse, BSN]

On 3/4/2020 at 10:13 AM, 42pines said:

Correct, but misleading. Simply stated: A case fatality rate of 3.41% stands only if every single case where the afflicted has not recovered, or died, will recover.

In other words, you are assuming that all remaining cases with unknown outcome will have no fatalities.

Nonetheless, my article's point is simply the "death" rate of the seasonal influenza over eight years is about one death in 750-1000 cases.

Using the 3.41% you have quoted means 34 deaths per 1,000 cases.

That means a death rate of about 34x that of the seasonal influenza.

Thus it is unwise for nurses to say: "oh, it's about the same as the flu."

On the other hand, the seasonal influenza has killed more this year already than has Covid-19 and we can lower that rate by immunization. Both are important subjects for nurses.

I'm in no way comparing this to influenza, but during flu season, I've seen so many providers prescribe Tamiflu without bothering to lab test the patient first. What about them? They aren't swabbed, so are they officially diagnosed with influenza? Are we counting positive flu swabs only? There are many + flu patients who never went to the doctor.

[myoglobin, ASN, BSN, MSN]

One of the best lines of evidence for a CFR of one percent or less is from the Diamond Princess Cruise ship as examined in this article https://www.sciencenews.org/article/coronavirus-outbreak-diamond-princess-cruise-ship-death-rate . This was a "closed system" where everyone was tested and followed intensely and their were six deaths out of around 700 infections. When the authors adjusted for the "older ages" of the cruise ship members (relative to the general population) they concluded it was likely that the actual infection fatality rate was closer to 0.5 of 1%. Again, if more tests were available we would likely see a case fatality rate that was much lower than WHO figures.

[Kyrshamarks, BSN, RN]

OK, so let's say the death rate is only 1%. I place a bowl of 100 M&M's in front of you and tell you that one is actually cyanide and will kill you. Would you take the chance and eat one?

[myoglobin, ASN, BSN, MSN]

Of course not, but it is important to try and understand real risk levels so that we can make the most informed decisions about trade offs and risk to benefit analysis as we move forward. Thus as a healthy 51 year old I am considering exiting my home based psych tele psych NP position (where I am only getting about four clients per day) and going back to the ICU where I used to work (and can get $50.00 12 hour shifts all month long). For me the benefit of not going more delinquent on my bills may outweigh the risk of disease (which frankly I was probably exposed to six weeks ago in Seattle when many around me were sick with some sort of flu like illness). Of course to do this I would need to get a test to show that I am currently negative so as to not expose patients (potentially).