Using statistical measures, you don't need to be able to point at any one person and say "air pollution killed this person". There are obviously other correlat

Using statistical measures, you don't need to be able to point at any one person and say "air pollution killed this person". There are obviously other correlations you need to be able to account for, but generally you can account for them. >It is not the same as saying that someone died of covid, like stated in the article. Interestingly, the covid death rate is largely calculated the same way. Someone doesn't have to die from acute respiratory issues caused by a covid infection in order to count, we just see that populations with high rates of covid infection also have higher rates of death. This helps also account for e.g. covid causing heart attacks which would otherwise be hard to account for.