'An epidemiological mystery': Why Is COVID-19 so much deadlier in some countries than others?

With the number of cases worldwide approaching 200,000, and 1,000 or more cases in 15 countries, you’d think there would be an answer. But the more data we see, the tougher it is to come up with a hard number.

What's the COVID-19 death rate? I'll get back to you

A patient in a biocontainment unit is carried on a stretcher at the Columbus Covid 2 Hospital in Rome, Monday, March 16, 2020. (Alessandra Tarantino/The Associated Press)

Imagine you're a chief public health officer and you're asked the question on everyone's mind: how deadly is the COVID-19 outbreak?

With the number of cases worldwide approaching 200,000, and 1,000 or more cases in 15 countries, you'd think there would be an answer. But the more data we see, the tougher it is to come up with a hard number.

Overall, the death rate is around four per cent — of reported cases. That's also the death rate in China, which to date accounts for just under half the total number of global cases.

China is the only country where a) the outcome of almost all cases is known (85 per cent have recovered), and b) the spread has been stopped (numbers plateaued about a month ago). 

A four per cent death rate is pretty high — about 40 times more deadly than seasonal flu — but no experts believe that is the death rate. The latest estimate is that it is around 1.5 per cent. Other models suggest that it may be somewhat lower. 

The true rate can be known only if every case is known and confirmed by testing — including the asymptomatic or relatively benign cases, which comprise 80 per cent or more of the total — and all cases have run their course (people have either recovered or died). Aside from those in China, almost all cases identified are still active. 

Unless a jurisdiction systematically tests a large random sample of its population, we may never know the true rate of infection or the real death rate. 

Yet for all this unavoidable uncertainty, it is still odd that the rates vary so widely by country.

Extreme differences

The new epicentre of the outbreak is in Europe, so let's focus there.

The graph below shows the crude (not adjusted for age) death rates in the ten countries with the most reported cases (all 1,000 or higher). 

The COVID-19 death rate varies greatly in the 10 countries with the most reported cases. (Data from by Steven Lewis)

Let's look at the European extremes: Italy and Germany. Italy has the most cases (28,000) but Germany has more than 7,000 and its number is growing rapidly. Yet Italy's death rate so far is 35 times higher than Germany's. 

Moreover, only two of Germany's 7,200 active cases are classified as serious or critical, while in Italy it's 1,850 out of 21,000. In Canada there have been four deaths among 422 cases and only one of the 407 active cases is serious or critical.

What's going on here? No one has a ready answer. 

The health care infrastructure in Italy and Germany is pretty much the same. Germany may be a week or two behind Italy in the pandemic, but in most countries the death rates seem higher at the beginning of the upsurge because the more serious cases are likely to present and be tested first. 

Germany has promoted widespread testing, which is more likely to confirm non-serious cases and prevent them from coming into contact with more vulnerable groups, but so has Italy. And Italy took stronger measures to contain the virus than Germany.

Death rates are much higher in the elderly — as much as 15 per cent in the 80+ age group – and Italy has a relatively old population, but Germany has the oldest median age in Europe. 

Does the German gene pool create some sort of uniquely effective immune response to just this novel coronavirus? Highly unlikely. Is the virus somehow different in Germany? Virtually impossible. Could German health care be that much more effective? Again, no. 

Then what?

At this point the discrepancies seem both massive and unexplained. If they converge in the next few weeks we will be closer to an explanation. It may be that Germany has thus far been lucky, with the exposures having occurred mainly among younger and healthier people. Or maybe Italy has been unlucky, with high rates of exposure in people with pre-existing health conditions that make them vulnerable. 

If the perplexing differences remain after a few more weeks and tens of thousands — or millions — of new cases, we will have a barnburner of an epidemiological mystery on our hands. 

The most revealing data over time will be the death rates in the oldest population groups, because the numbers are higher and the statistics will be more reliable. If these still vary hugely, a lot of analysts will be scratching a lot of heads. 

Where is Canada headed? At this early stage, it looks more like Germany than Italy, but that can change quickly. 

Germany may not look like Germany on April 1. With any luck, Italy may not look quite so dire by then. To channel Donald Rumsfeld, there are unknown unknowns. For now, we are stuck with a paradox on either side of the Alps. 

What's the COVID-19 death rate? I'll get back to you.

This column is part of CBC's Opinion section. For more information about this section, please read this editor's blog and our FAQ.

Interested in writing for us? We accept pitches for opinion and point-of-view pieces from Saskatchewan residents who want to share their thoughts on the news of the day, issues affecting their community or who have a compelling personal story to share. No need to be a professional writer!

Read more about what we're looking for here, then email with your idea.


Steven Lewis

Health policy analyst

Steven Lewis is a health policy consultant formerly based in Saskatchewan. He now lives in Vancouver.