From https://en.wikipedia.org/wiki/List_of_the_verified_oldest_pe... there are 43 people who reached 115, and we know how old 40 of them turned out to be. 9 of those made it to 117. That fits pretty well with the 50% odds of dying each year. She supposedly lived about 7.5 more years after that point. What would her odds be? 0.5^7.5 = 0.0055242717280199 are pretty bad. But with a pool of 40, the odds that SOMEONE lived that long turn are 40x that, which is around 22%.
When you combine worse healthcare, challenges around record keeping, and specific evidence of fraud, I am inclined to believe that she was fraud. But the statistical case for it isn't as overwhelming as I would naively guess.
(That said, access to a larger data set could easily make the odds against her existence much more extreme.)
https://www.populationpyramid.net/
But even at that scale you'd expect there to be one or two people reach that age and the article claims this case is actually well outside any of the statistical models. So the 50/50 statistic probably understates the odds as I'm sure those percentages keep getting worse every year.
1 - https://www.washingtonpost.com/news/speaking-of-science/wp/2...