I believe the 100 (or any of the numbers in that chart) are scores which he normalized to 100 in each age category for 1968. I don't know what the per 100,000 is supposed to mean. As a death rate per year, it doesn't seem possible. Consider that it's showing ~70/100,000 in the 85+ category. Given that they'd all be dead in 25 years, that number can't be right.
