The black fraction of the population of UK was never mentioned. If you're going to compare the life span of two places, there doesn't seem to be any reason to bring race up. If you're going to bring it up, you need to justify why its relevant. You can probably find lots of demographic lines along which you can split a population to support this argument or that argument. Some fraction of Mississippi is black, as is some fraction of the UK. And each sub-demographic has some life expectancy. Different places have different population mixes, but those mixes are de-facto representative of those places. If the argument is about non-black life spans, your argument would make sense. But if it's about the average lifespan of the region, and the demographic mix is different, it's non-sensical to filter using different cohorts since that mixture difference is a real difference between the regions.