As we are nearing the expected peak of the COVID-19 pandemic, our clients are looking for individualized answers on how deadly the coronavirus is for them. We now have more data coming from around the world as well as anecdotal first-hand evidence. This information sheds new light on the lethality of the virus. Here is our summary and conclusions from data gathered and analyses performed on what it means for you:
- Officially reported fatality rates are expressed as deaths divided by known (officially registered) number of infected patients (this ratio is called Case Fatality Rate or CFR). This ratio currently ranges from 6% in China to 13% in Italy and, according to many scientists, is overstating the true lethality of coronavirus. This is due to the fact that only a small proportion of all infected are officially captured by country statistics due to the shortage of tests as well as the fact that more than 50% of infected individuals do not show any symptoms. Therefore, the CFR ratio has the correct nominator, but the incorrect denominator.
- A better statistic for evaluating lethality of the virus is Infection Fatality Rate (IFR), which measures the number of deaths compared to all actual infections (not just officially registered ones).
- So far, our best chances of capturing IFR were onboard the Diamond Princess cruise ship (where over 3,000 out of 3,711 people onboard were tested), as well from the experience of Iceland, a small country, which tested an unusually high proportion of its population (approximately 13%) and not just the symptomatic individuals. Also, large studies of asymptomatic blood donors in Sweden have been very useful to capture what percentage of the general population already has the virus.
- The IFR, based upon analysis of the Diamond Princess and Oxford’s CEBM Research study published last week, is approximately 0.5%.
- John Ioannidis at Stanford University estimates the IFR in the U.S. at 0.125%.
- The latest estimates of the expected total death toll for the U.S. is 60,000, which implies IFR at 0.05-0.12%.
- Additionally, the news from Sweden as the only developed and sizeable country that didn’t introduce a lock-down, are encouraging. Karolinska Institute and Tom Britton studies suggests that 11-40% of Stockholm population already had the virus. The death total so far is around 1.7 thousand in the whole country of 10 million people. The Public Health Agency predicts that by May 1st, 30% of the Swedish population will already have had the virus, which means that its true lethality is much lower than anticipated.
- Last weekend, UCSF launched an effort to conduct a similar study, ie. testing whole segments of population of a particular area (in their case Mission district of San Francisco, testing 5,700 people) to find out what the real infection rate is. We will monitor this closely.
- On the basis of all those studies, we can provide broad ranges for adjusted probability of dying from the virus if infected. The pessimistic adjusted ratio will be the higher end of the range announced by the abovementioned institutions (0.5%) and the optimistic adjusted ratio will be the lower end of the mentioned institutions (0.1%).
- When we allocate the distribution of deaths for IFR by age group according to known CFR rates published by Chinese CDC for the data up to 11thFebruary, the probability of dying from coronavirus in a particular age group for an infected person is as follows:
* Adjusted figures reflect much larger denominator, ie. expected much higher amount of infected, who go unnoticed due to lack of test kits or lack of symptoms.
- Zunyou Wu, MD, PhD1; Jennifer M. McGoogan, PhD1, Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China; JAMA, 2020;323(13):1239-1242
Your Healthdom Team