COVID’s Changing Numbers
The infection fatality rate is the number of deaths divided by the number of infections. Early 2020 estimates of the IFR for COVID, for example, ranged from 0.17 to 1.7% and are considered high. John Ioannidis and others recently reported a median IFR of 0.095% for the 0 to 69-year-old population. This was based on data from 29 countries and included publicly available COVID death data and age-stratified seroprevalence information.
Also, several studies published last year have suggested that the death rate of COVID may be similar to the death rate of the flu. These include a study in the International Journal of Infectious Diseases and another in the International Journal of Epidemiology.
Likewise, early estimates of the case fatality rate (CFR) for COVID – deaths divided by the number of cases measured by positive tests – were high. A March 2020 assessment of the CFR in The Wall Street Journal questioned initial estimates of 2 to 4% as too high by orders of magnitude.
Why the changing numbers? Some possible explanations (there are surely others):
- Initial inadequate counting led to varying denominators in the IFR and CFR estimates.
- Varying contagiousness of the disease.
- Varying case rate (number of infections divided by the population during a period).
- The initial outbreak in institutionalized elderly and frail people (living in nursing homes/care homes) before turning toward others in the community.
Thus, there was increasing evidence by the summer of 2020 that COVID infection and case fatality rates were extremely low for people under 70. Public health officials should have relied on previous experience and modified their policies towards more targeted strategies – protecting the elderly and frail and indicating there is no longer a reason to use large-scale restrictions (e.g., stay-at-home, masking orders) on the young and strong.
Some risk stratification was attempted. Because the initial supply of vaccines was limited in late 2020, officials recommended which groups should receive the earliest vaccine allocations. This included healthcare personnel, other essential workers, institutionalized adults with high-risk medical conditions, and adults aged 65 and over.
Risk stratification was short-lived, and the logic for other public health interventions was far less clear. When large-scale restrictions were imposed, certain essential businesses, e.g., grocery stores and pharmacies, were kept open; other “non-essential” businesses like gyms and theatres, were shuttered. Many decisions came across as arbitrary. How is the local GNC essential?
Protecting the Frail
Among the elderly and frail, death came to those with comorbidities more frequently than those without additional health problems. Of 1,738 COVID hospital patients dying in Italy by April 2020, for whom it was possible to review clinical charts, 96.4% had comorbidities. Early work in New York City pointed to age, heart failure, male sex, chronic kidney disease, and obesity as significant comorbidities.
Additionally, the linkage of comorbidity to COVID infections and deaths is not normalized to the population. Hypertension is a good example of the need for this, as it is one of the most prevalent health conditions in the elderly population.
For example, hypertension was identified as the most common comorbidity in those hospitalized for COVID. But when that number was adjusted for hypertension prevalence, it had a much-diminished influence on hospitalization. And measures of co-morbidities may be further confounded in analysis because they are commonly found in combination in people over 65, including hypertension and coronary artery disease, cerebrovascular disease, and diabetes.
So, by summer 2020, it was increasingly apparent that the “healthy” elderly were at low risk. There was no reason to continue to frighten these people.
The institutionalized elderly and frail fared poorly in the pandemic. This is not a unique feature of COVID. Harvesting, or mortality displacement, is a term applied to an infection accelerating deaths forward in time that, without the infection, would occur later anyway. Each year, more people die in winter than in summer, with seasonal flu, among other factors harvesting the elderly and frail.
A perfect example is recent research into excess winter mortality fluctuations in Western European countries due to the influenza A virus subtype H3N2. During good years, without influenza A/H3N2 outbreaks, mortality is very low, and the elderly and frail are allowed to “accumulate.” During bad years, with an influenza A/H3N2 outbreak, the elderly and frail are “harvested,” dramatically increasing the number of deaths.
In Italy, the average or median death age of those dying with COVID was approximately the same as those without, ~80 years old. In England and Wales, the median age of those dying with COVID and those due to COVID was the same, 83 years old.
The word death is harsh and misleading when it implies that there is one uniform case fatality rate. Much of the messaging by public health officials about COVID fatalities failed (purposely?) to draw attention to the age or frailty factor. The reality is that there were few, if any, years of life lost for the elderly and frail due to COVID. For most healthy people, COVID is silent or unimportant.
Low infection and case fatality rates were known by the summer of 2020. Right here, public health officials should have modified their policies toward more targeted strategies. Failure to do so, i.e., maintaining large-scale restrictions, was a big mistake with disastrous consequences on the American economy and society.
Let’s be clear. Public health officials should have followed the evidence. Others will be left to sort out how their failure to act in the public's interest came about.
