Last updated: 27 March 2020
A confirmed case is, or should be, “a person with laboratory confirmation of COVID-19 infection” (as defined in WHO situation reports); but different countries apply different standards, and standards have changed over time in some countries. Confirmed cases are typically reported both on a daily and on a cumulative basis. The cumulative figure therefore also includes people who went on to recover, as well as those who went on to die, and therefore is not equal to the number of current cases.
The number of confirmed cases is used to calculate the widely-discussed Case Fatality Ratio (CFR), which is simply the number of deaths attributed to COVID-19 divided by the number of confirmed cases.
For many reasons, we should care more about the number of actual cases, whether confirmed or not.
And similarly, we ultimately care about the Infection Fatality Ratio (IFR) more than the CFR, which is calculated by dividing the number of deaths attributed to COVID-19 by the number of actual cases.
Why? When you read about the estimated final attack rate (the proportion of a given population — a country, the whole world, a family, a cruise ship — who become infected over time), and want to translate that into potential fatalities, you need to apply the IFR rather than the CFR.
That’s because estimates of the attack rate are estimates of how many people will actually get COVID-19, not how many cases are detected.
For a much more detailed (and, frankly, better) discussion of this, see this section in the always excellent Our World in Data. Their section also includes a discussion on important, related topics which I haven’t had time to write about:
- Why CFR is not a constant over time
- Why CFR does not represent the risk of death for an infected individual
- Examples of how estimates of CFR have declined over time
Another interesting source on this last topic is a newsletter published by JP Morgan in early March. The relevant section:
The complex thing about pandemics is that early mortality rate estimates tend to decline over time. Why? Here are four simple measures that matter in the context of a pandemic: (a) the population of a given geographical area (b) the total number of infected individuals, including both asymptomatic people and people that get sick (c) the total number of people that are infected, get sick and self-report (d) the total number of people that die During the haze of a pandemic, the best estimates that entities like the World Health Organization often derive are based on (a), (c) and (d), and even things like (d) are complicated by pandemics affecting older individuals with pre-existing conditions. They do not know (b) upfront, and sometimes it is never known, or only known with the passage of time.
Take the Swine Flu (H1N1/2009) as an example. Early estimates in the fall of 2009 from the WHO1 pegged the H1N1 mortality rate at 1.0%-1.3%, since they were dividing (d) by (c). Four years later, a study from the WHO and the Imperial College of London2 estimated H1N1 mortality as a function of total infections, including both the asymptomatic and the sick. Their revised H1N1 mortality rate using (b) as a denominator: just 0.02%.
So, please treat estimated infection rates and mortality rates with care, since they can mean very different things. Marc Lipsitch from Harvard has estimated that 40% – 70% of the world’s population could become infected . Lipsitch himself makes it clear that this number is an example of (b) and not (c) and that there is
an enormous gap between the two, so please do not multiply population by 40%-70% and then multiply by a mortality rate assumption. The vast majority of infected people will likely not become sick, and around 80% of people who get sick develop mild infections rather than severe ones.
The Our World in Data section I linked to above illustrates just how quickly our understanding of CFR can evolve:
What are our best current estimates of CFR / IFR?
The epidemiologist Adam Kucharski, a great source, points to three papers that estimate the range in China at 1-1.5% of symptomatic cases (importantly that means CFR, not IFR):
Subjective Probability 
3 thoughts on “What are IFR, CFR, and attack rate?”