If I have no known health problems, how does that change the distribution? Are there richer datasets that include that type of breakdown?

From a different angle, if 15% of males my age will die in the next 30 years, how many of those are foreseeable due to current circumstances vs accidents that could happen to anyone?