The Infection fatality rate (IFR) has been debated since the beginning of the pandemic. This number represents the percentage of people who die out of the total number of people who were infected with the virus. We recall that the initial modeling study out of the Imperial College of London suggested an IFR of 3.4%. This figure was incorrect and inconsistent with the best available case data at the time, but unfortunately, this modeling study dominated and misdirected public health policy in many countries.
In March, the raw case fatality data from South Korea demonstrated that those under the age of 60 had less than 0.2% risk of dying if infected with COVID-19, and there were no deaths in thousands of people under the age of 30 who tested positive. As of 10/25/20, South Korea’s case fatality rate for those under the age of 60 is 0.15% (27 deaths out of 18,517 cases) and there continues to be no deaths under the age of 30 due to COVID.
Until 9/10/20, the CDC suggested infection fatality rates of 2.4%, 1%, and 0.6% based upon their modeling estimates. In March, Dr. John Ioannides, an epidemiologist and infectious disease physician from Stanford suggested the IFR was between 0.05 to 1% based upon actual cruise line case data. In April, he coauthored a serology research study in Santa Clara county that was consistent with other serology data to suggest the disease was much less lethal than suggested by the CDC. In May, he submitted research of the prevalence of the virus and IFR from world-wide serology data. This global research data demonstrated a median IFR under the age of 70 of 0.05%. His important research lingered in preprint status until September 15, 2020 when it was finally peer reviewed and accepted for publication as a WHO bulletin article. The link below to this research article that was displayed online October 14, four months after it was submitted for publication.
The CDC updated IFR values for modeling the COVID-19 impact on the U.S. on September 10, 2020 are displayed below.
The last column to the right is the CDC’s best estimate for the number of people who will die if they are infected with COVID based upon their age demographic. They only included those under the age of 80 for these calculations citing incomplete data in those over age 80. Breaking down the numbers above into more understandable terms, results in the following:
0-19 years: 1 in 33,333 chance of dying if infected or a survival rate of 99.997%
20-49 years: 1 in 5000 chance of dying if infected or a survival rate of 99.98%
50-69 years: 1 in 200 chance of dying if infected or a survival rate of 99.5%
70-80 years: 1 in 18 chance of dying if infected or a survival rate of 94.6%
HOW LONG DOES COVID IMMUNITY LAST?
The link below is a September research article from the New England Journal of Medicine. The study demonstrated detectible levels of IgG antibodies 4 months after an infection. Seroconversion occurred in over 91% of individuals confirmed to have COVID by PCR testing. The researchers were using two pan-immunoglobulin assays that are not available commercially. There is another study on the CDC website that replicates this research of persistent humoral immunity for nearly 3 months. Additionally, there is now consistent research evidence that T-cell immunity is impacting population susceptibility as well.
RESEARCH DATA ON EXCESS DEATHS AND THE IMPLICATIONS FOR DISEASE PREVALENCE
An article in the October 20th edition of JAMA entitled Excess Deaths From COVID-19 and Other Causes March-July 2020, identified some interesting questions by measuring the number of unexpected deaths between March 1st and July 31st. The Oregon data suggested that there were 1123 unexpected deaths but only 338 documented COVID-19 related deaths during this same time period. This finding would suggest that COVID-19 represented only 30% of the unexpected deaths. This data implies that COVID deaths were underrecognized or there were excess deaths due to non COVID related disease. Both scenarios likely occurred in my opinion. Evergreen Family Medicine experienced large volumes of patients with severe respiratory conditions in the urgent care and hospital in February and March while testing was not available. If all or most of these deaths were due to COVID, then the prevalence of the disease is more than public health officials have acknowledged.
As of 10/25/20, Oregon reports 653 deaths out of 42,101 cases. If we use the commonly cited assumption that there are 10 times more cases than are measured (an underestimate of the number of cases in my opinion), then 420,000 Oregonians have had COVID. If we assume that all the excess deaths from March 1st through July 31st were due to COVID (unlikely), assuming the same ratio of cases to deaths (also unlikely given the apparent decreased infection lethality trends), there would be an additional 500,000 Oregonians who have been infected. There are 4.2 million people in Oregon. Using the conservative assumptions above, nearly 25% of Oregonians have had the disease.
If the excess deaths are due to other non COVID conditions, what are those conditions and why is dying of one disease worse than another? Are there increased numbers of deaths in younger people because of public health policies and not due to the virus itself? Sometimes the most important questions are not asked enough, or unfortunately, deemed by some as unethical and irresponsible to ask at all.
There are many reasons why I have recommended a Focused Protection approach during this pandemic. It is only reinforced by the evidence above. Contact tracing children and otherwise healthy people is not a focused approach and is a misallocation of resources. The link below is an article from Dr. Ioannidis that appeared in the European Journal of Clinical Investigation from 10/7/2020 that provides an important perspective on COVID-19 epidemiology.
I hope the information is helpful and provides a realistic measure of hope as people navigate their own risk tolerance going forward.
John Powell M.D.