There's also a plausible hypothesis that there's a selection bias in effect when it comes to robustness in women.

The balance of births in a normal population ever so slightly favours men. That is, more boys are typically born than girls.

And when you control for all other factors, it turns out that pregnancies involving girls are slightly more likely to result in complications resulting in foetal death or spontaneous abortion. We don't know why.

What does this have to do with life expectancy? Well if girls have a tougher time getting through gestation, then that's an evolutionary selection pressure. Girls need a little "something" extra to survive the process than boys do.

As a result, the hypothesis suggests, born females have a biology which makes them slightly more resilient than born males. This makes little difference through most of life, but as you approach the end of life and the great longevity filter, those who are more robust will survive longer. Which happens to be women. Resulting in these huge demographic disparities when you get to the extreme ages.