COVID-19 is much worse than the flu.

With influenza epidemics the situation seems quite unlike what’s happening with the SARS-CoV2 pandemic.

In most modern communities there is substantial pre-existing community protection against infection because of immunological memory from vaccination and previous influenza infections.

Penetration of influenza infections into the community in flu epidemics has been be more limited than we are seeing with Covid-19 (with flu attack rates of 10 to 30% being quoted by various sources). But rigorous surveys of influenza attack rates during epidemics are few and far between.

These differences mean that even if influenza and Covid 19 had identical infection fatality rates differences in attack rates would make Covid-19 more dangerous because almost all people can be infected.

Measuring fatality rates isn’t easy.

Infection fatality rates and case fatality rates are not easy to measure accurately for either virus.

It might seems obvious and simple to compare the two diseases by working out what percentage of infections result in death, and thus obtain accurate values  the metric called infection fatality rate (IFR).

A problem is that we don’t have usually have measurements of the numbers of people who are infected. It’s expensive and difficult to get this number. Infected people can’t always be counted because many don’t look or behave differently from non-infected people. The surprise with Covid-19 in the early stages of the pandemic was that many people were infected without realising it. They were “asymptomatic’ as the medicos say. This made the epidemic very dangerous and difficult to control. It also makes infection fatality rates difficult to pin down.

But what is not always realised is that the same kind of under-the-radar spread can occur with influenza.

If estimates of Covid-19 infection fatality rate have been a bit rubbery, flu IFRs are equally rubbery.

It’s sobering to read a systematic review of this in the context of the 2009 influenza pandemic in the Wong and colleagues Epidemiology review:

Estimates of deaths ranged from 1 to more than 10,000 per 100,000 affected people for this strain of flu. To make it even more confusing, several different criteria for identifying infection numbers or influenza cases were used by different investigators, adding to spread in the estimates of rates.

Wong and crew concluded:

Conclusions: 

Our review highlights the difficulty in estimating the seriousness of infection with a novel influenza virus using the case fatality risk. In addition, substantial variability in age-specific estimates complicates the interpretation of the overall case fatality risk and comparisons among populations. A consensus is needed on how to define and measure the seriousness of infection before the next pandemic.

IFR for flu 2009 was at most, 10/100,000 people infected.

When influenza fatality rates estimates were based on a denominator where numbers of infections were estimated, infection fatality rates in the 2009 flu pandemic ranged from 1 to 10 deaths per 100,000 infections.

It is noteworthy that most, if not all, Covid-19 infection fatality rates are much higher than this, typically being 500 deaths per hundred thousand people in well developed countries.

Many people have noticed that the worldwide IFR value for the coronavirus pandemic is lower than this (say 150 per 100,000, but still well above 10 per 100,000 shown by flu). The same people do not always factor in the uncertainties in this global IFR value, or realise that it’s influenced by the age distribution of the communities affected. Young countries, such as many places in Africa have much lower COVID-19 infection fatality rates than countries such as Italy with an ageing population.

Total death numbers provide a metric that is independent of uncertainties about infection numbers.

There is a way round these confusions about infection fatality rates—count up the total numbers of deaths from two different epidemics and compare flu and COVID-19 death numbers directly.

There are reliable numbers for total death numbers Covid-19 and alternative ways of getting those numbers to crosscheck the values.

One of these is the so-called excess deaths measure that is the numbers of deaths above the expected number of weekly deaths can be used to identify whether unusual causes of high death rates have occurred. The graph below captures deaths for flu and COVID-19 in the one diagram.

The graph above shows a plot produced by the US CDC to display weekly numbers of deaths from all causes in the United States. From previous years stats, the expected profile of deaths can be estimated for each week, as shown by the orange line that shows the upper bound for normal expected deaths numbers for each week of the year. For weeks where the counted death numbers  significantly exceed this upper bound there is a cross. 2017-2018 was a bad year for influenza as shown by the 8 weeks with high death numbers or “excess deaths”.

Other calculations showed that the US deaths resulting from the 2017 to 2018 influenza pandemic caused by the flu virus were 61,000. (For example see https://www.cdc.gov/flu/about/burden-averted/2017-2018.htm).

In contrast to 2017-2018, excess deaths in 2020 – 2021 were much greater than during the influenza pandemic of 2018 and extended over a much longer interval as shown the row of crosses in the graph. These excess deaths were mostly caused by SARS-CoV2, as there was essentially no influenza circulating in the USA.

We don’t know the final death numbers for the COVID-19 epidemic yet, because it’s not yet over.

But just a glance at the above CDC graph show that they will much, much greater than the 61,000 souls dying from flu  in 2017-2018. At the time of writing (July 2021), in the USA, COVID-19 deaths are estimated by the CDC as 595,387 , so that flu and COVID clearly differ in lethality by a factor of ten-fold or so based on death counts. The previously mentioned infection fatality rate estimates we do have show a similar pattern too.

COVID-19 is simply not just the flu.