Thanks for your thoughtful comment.

I have to correct your “Tsimane average life expectancy” argument. First, average life expectancy is the wrong metric to apply. Here is why: Average life expectancy of a population is calculated over all life durations, including pediatric deaths. An illustrative example: if one person lives to 60 and another to 40, their average life expectancy is 50. If one lives to 60 and the other one dies as an infant within a year of having been born, the average life expectancy is 30.5 years. In populations like the Tsimane, infant mortality is far higher than in the US, which drags down the average. This is why median life expectancy or life expectancy at a specific age provides a more accurate picture of a population's health. This point is overlooked in many articles that present their arguments on the basis of average life expectancy. In the case of the Tsimane, the life expectancy at 50 is comparable to the US.

Second, the comparison that I made in the post explicitly refers to that age group.

The case that you present is certainly an interesting one, but as anecdotal evidence goes, it is not proof of anything. As I mentioned in my response to Georgios, diseases and biomarkers exist on a probability distribution. In biology it is inevitable to encounter extreme cases that reside at one of the tail-ends of the distribution. They are proof of the distribution, but never of the majority of cases. That’s why we have the concept of risk. Even if you live the healthiest life imaginable, there remains some residual risk. One more illustrative example: George Burns lived and acted on stage until he was 99 years old. He still smoked several cigars a day at that age. And he didn’t even die of any of the consequences of smoking (he died of the consequences of a bad fall in the bathroom). Given his anecdotal case, would you recommend dismissing smoking as a threat to health? I don’t think so.