Italy's coronavirus death toll could be 'substantially' higher than official figures show
15 May 2020, 00:55
Italy's official coronavirus death toll could be a "substantial underestimation" of the actual number of people to die with the disease, according to a study.
Researchers studied the city of Nembro in Lombardy, one of Italy's worst-hit regions, and worked out the change over time in the total number of deaths - known as all-cause mortality.
The results showed that more people died in the city in March 2020 alone than in the whole of 2019, or any of the previous seven years.
Official figures showed that roughly half the deaths to have occurred there during the recent coronavirus outbreak were reported as confirmed Covid-19 deaths.
For example, a total of 166 deaths were recorded in the city between 21 February and 11 April this year. In that same period, 218 people tested positive for the virus, of which 85 died.
This suggests that 81 people died of non-Covid causes during that time.
However, deaths in Nembro for that period would usually be far lower, which suggests some of those 81 people did in fact have the disease, according to the study's authors.
If so, and applied to the rest of the country, it would mean that Italy's official death toll is far higher, perhaps even greater than the UK's.
Between January 2012 and February 2020, Nembro's all-cause mortality rate stood at roughly 10 per 1,000 person-years, peaking at 21.5 per 1,000 person-years.
One person-year accounts for the number of people in a study and the amount of time each person spends in the study. Therefore, if 1,000 people were studied for exactly one year, that would leave you with 1,000 person-years of data.
But in March, the all-cause mortality rate rose to a peak of 154.4 per 1,000 person-years. The previous year, the same month had a rate of 14.3.
The researchers said: "Across Italian cities, all-cause mortality has notably increased because of the Covid-19 pandemic but this increase is not being completely captured by officially reported statistics on confirmed Covid-19 deaths."
They suggested the discrepancies could be due to shortages of tests to confirm the virus, or patients dying from causes indirectly related to the disease, such as limitations on healthcare capacity due to the pandemic.
The authors added: "Our findings imply that the reporting of confirmed Covid-19 specific deaths represents - at least for some Italian regions - a substantial underestimation of the actual number of deaths from the disease."
They concluded: "The results suggest that the full implications of the Covid-19 pandemic can only be completely understood if, in addition to confirmed deaths related to Covid-19, consideration is also given to all-cause mortality in a given region and time frame."
The peer-reviewed study, by experts at the Institute of Public Health and Charite-Universitatsmedizin Berlin and published in the BMJ medical journal, based its findings on the monthly all-cause mortality data from Nembro between January 2012 and April 2020.
It also took into account the number of confirmed deaths from Covid-19 up to April 11 this year and the weekly absolute number of deaths between January 1 and April 4 across recent years by age group and sex.
However, because it is a descriptive study, it cannot formally establish a causal relationship. The authors added that some of the data used might be provisional.
They said the steep increase in all-cause mortality was even more pronounced after further analysis to test the robustness of the results.
Their findings back up results from a recent larger report from more than 1,000 Italian cities.