Take a look at the beta version of dw.com. We're not done yet! Your opinion can help us make it better.
Comparing coronavirus statistics is virtually impossible right now because each country collects the data differently. DW asks public health expert Professor Tobias Kurth of Berlin's Charité hospital to explain why.
DW: We read statistics on infection and death rates from the coronavirus in Germany every day. And we get those statistics from countries around the world. To what extent can these numbers be trusted? Are they accurate?
Prof. Tobias Kurth: The numbers [we have in Germany] are accurate. The question is, however, whether you can compare them to the numbers from another country. And that is very difficult, as other countries are reporting their numbers differently.
They are testing differently, or they are classifying people who died from coronavirus or the associated symptoms differently. So it's not as though the numbers that we report in Germany aren't right. It's just that as soon as you start to compare those numbers to another country — for example, Italy — it gets very challenging.
Could you expand on that and give us an example of a comparison with another country?
Yes. So, for example, if you test a different group of people for the coronavirus [than in another country], you will have a different population who potentially can test positive for the virus. In Germany, the average age of people testing positive is about 45. In Italy, it's over 60 years of age.
So in Germany, we have a larger group of people who are younger who test positive. But that also means that the group in Italy is older, and we know that older age is a high-risk factor for dying from the coronavirus infection, and the average age of people who die is the same — slightly over 80 in both countries — so one should not be surprised that when you look at the absolute numbers that are recorded, Italy has a larger number compared to Germany.
There is another factor that may not be relevant right now, but the difference in the death rates in different countries, for example, in Italy and Germany, might also be due to certain social structures. We know that the elderly are better integrated into social life in Italy compared with Germany, and that is usually a good thing.
But in this case, you increase the likelihood of getting infected, and the elderly are those who are more likely to die from this disease. So that may also explain some of the differences that we're seeing compared to Germany, where the elderly are maybe a little more isolated than those living in Italy.
And do we know how many deaths normally occur around this time of year? Do we know what a normal number is compared to what we're seeing now?
That is a very good question and a very important question. And these numbers are actually very much needed to see if there really is an increase in overall mortality. Reports from Italy suggest, at least in some regions, that the number of people who are dying right now and [are] not classified as coronavirus patients is increasing as well. So it is likely that for each coronavirus death there is another person who died because of coronavirus but was not recorded as such.
But it depends on the country whether you have data available on how many people died last week, for example, so that you can compare that to an average rate of death in that week for the past five years, let's say. If there's an increase, you would then potentially see a signal that something was going wrong.
Only a few states report to a European database so you can see these numbers fairly quickly. And to the best of my knowledge, we do not have that data available in Germany right now.
So there's not really much being done to compare those numbers with deaths from Covid-19 because the data's not there?
Not in Germany. But I'm aware of reports from Italy. They do exactly that. They compare the number of people who died in a given week — let's say, last week — to the deaths in that calendar week of a previous year, or an average of the previous five years. And there, you can see a clear peak.
How likely is it that we will see a huge and perhaps an unexpected spike in deaths in Germany, or even elsewhere, because not all coronavirus deaths are being recorded as coronavirus deaths?
In Germany, I think we are going to see more people die from the coronavirus. And we will know sometime in the future whether there is also an increase in the overall mortality, likely also caused by the coronavirus but not recorded at this point.
Globally, the same thing could be true, because we're not testing everyone. In addition, we don't test a person who dies [as a non-coronavirus patient] for whether she or he actually had a coronavirus infection. But that would allow us to classify people differently than we currently do. So, yes, there will be an excess rate of people who die from the coronavirus that we are currently not reporting and not seeing.
Professor Tobias Kurth is the director of the Institute of Public Health at the Charité University Hospital in Berlin.