A widely shared post on Facebook compares the number of deaths from diseases categorised as “flu & pneumonia” (which in practice means deaths from flu or pneumonia) in the first eight months of five different years.
It shows that far fewer people have died of these diseases so far this year, compared with 1976, 1982, 1997 or 2000. “Does anyone else find this just a little bit odd?” the post says, which some readers may understand to imply that tens of thousands of flu and pneumonia deaths are being wrongly recorded as deaths from Covid-19. (This is a common misunderstanding.)
In fact, this post is a perfect example of what’s sometimes called “cherry-picking”—choosing unusual examples to suit an argument, but hiding the whole picture.
What is wrong with the numbers?
The figures themselves are accurate. They come from data for England and Wales published by the Office for National Statistics (ONS) on 8 October 2020. What’s odd is the choice of years to look at.
As you can see from the graph below, there have been relatively few deaths from flu or pneumonia in the first eight months of 2020. It’s the lowest figure in recent decades, but not outlandishly low, when compared with the number of people dying in the same period of recent years.
By contrast, the years 1976, 1982, 1997 and 2000 are the four years with the highest flu and pneumonia deaths in the first eight months on record (going back to 1959).
And a fall in flu and pneumonia deaths might have been expected this year, since many of the social distancing and hygiene measures introduced to reduce the spread of Covid would be likely to reduce the spread of other infectious diseases as well.
Indeed, evidence from countries in the southern hemisphere (during what would typically be their winter flu season) suggests that this is exactly what happened, as there were significantly fewer flu infections than normal.
It’s also possible that some high risk people who might have died from flu or pneumonia in a normal year may have died from Covid instead in 2020.
If deaths from flu or pneumonia in the first eight months of 2020 had been similar to the numbers recorded in the four highlighted years, that would have been more surprising.