Looking at the numbers of reported cases and deaths from the new coronavirus in different countries, the percentage of people who catch the disease and eventually die (the fatality rate) appears to vary greatly. For example, as of 07 April 2020, Germany reported that 1.6% of 99,225 reported cases had died, while the UK reported 10% of 51,612 cases had died. Does this mean that the virus is for some reason more deadly in the UK than Germany?
Reading the numbers
To understand these figures, it is important to know that the number of COVID-19 tests carried out in each country varies greatly, and that many people with mild illness are untested and therefore unreported. Germany had performed 16 tests per 1000 people by 05 April 2020, while the UK had carried out 3 per 1000. This means the UK had not detected as many of the milder cases as Germany, and this is one reason why the numbers of reported deaths relative to cases is higher in the UK.
Furthermore, the time-lag between being infected with coronavirus and dying from COVID-19 can make the death rate appear lower than it really is in the early stages of a growing epidemic. The number of COVID-19 deaths on a given day depends on the number of people who were infected about three weeks earlier. Countries where the epidemic is just beginning, report new cases when relatively few of them have had the disease long enough to have died, so the fatality rate appears to be low.
Studies that have accounted for both the undetected milder cases and the time-lag until deaths are observed, estimate a COVID-19 fatality rate of around 0.7-1.1%. A more definitive figure will only be possible once antibody tests are used to find out the true percentage of the population that has been infected with the coronavirus, and when the outcomes of illness have been monitored and reported more carefully among people who become ill – in hospitals, in family and care homes, and elsewhere in the community.
Real differences in fatality
Aside from these complications in interpreting the data, is there any evidence that fatality rates really do differ among countries? It is too early to put exact numbers on this, but a number of factors are likely to cause real variation in the likelihood of dying from COVID-19 after being infected.
COVID-19 has a higher fatality rate in older people. One study among the many that have now been done found the risk of death in people who developed symptoms to be five times higher in the over-60s compared to those aged 30-59 years. The proportion of older people in the population varies between countries, for example 23% of the population of Italy is over 65 compared to 12% in China. This could help to explain why Italy has reported more deaths among COVID-19 cases than China – 12% vs 4%; but the data are not yet precise enough to tell definitively if this is the case.
Underlying health conditions, such as high blood pressure and diabetes, and heart and lung diseases, also increase the risk of developing severe COVID-19 disease, irrespective of age. For example, patients with coronary heart disease in China were twice as likely to die with COVID-19 as individuals of the same age without the condition. The prevalence of these underlying health conditions varies among communities within as well as between countries, and their contribution to COVID-19 deaths has yet to be ascertained.
The numbers of people in different age groups, and patterns of social contact between them, are also likely to be important. Drawing together data from different countries, one study estimated that the percentage of people dying from COVID-19 could vary seven-fold from 0.2% in Uganda to 1.4% in Japan, due to differences in social contact patterns and in the percentage of people in older age classes.
The quality of healthcare varies among countries too. For the minority of COVID-19 patients who experience severe illness, oxygen and mechanical ventilation saves lives. So far, most estimates of the fatality rate come from China which has a well-developed health system. Countries with lower availability of critical care might have higher fatality rates. The number of hospital beds per capita is around four times lower on average in low-income compared to high-income countries, and the percentage of these which are ICU (intensive care unit) beds is about two-fold lower in low-income countries on average.
In summary, there are multiple factors to consider when distinguishing between real and apparent differences in COVID-19 fatality rates. Different testing rates for COVID-19 and the stage of the epidemic are two of the reasons for apparent differences among countries. The age and contact patterns of people in society, the prevalence of underlying health conditions, and differences in the availability of healthcare are reasons why fatality rates might actually differ. In future, large-scale antibody testing to find out how many people have been infected, together with careful monitoring of the outcomes of illness in different population groups, will reveal the true differences in COVID-19 fatality rates among countries.