This study is comparing now to the early 1900s, when Aspirin was just starting to take off. Selection bias probably plays a very large part; they may have been

This study is comparing now to the early 1900s, when Aspirin was just starting to take off. Selection bias probably plays a very large part; they may have been able to measure temperatures but I doubt they were rigorous about taking representative cross-sections in 1910. Additionally the bias towards obesity has roughly switched- wealthier people were more likely to be obese in the 1900s, and now poorer people are more likely to be obese. Given that the people involved are in their 20s I would guess that obesity plays less of a role than you'd expect. [1]: https://en.wikipedia.org/wiki/Obesity_in_the_United_States#/media/File:Adult_male_obesity_in_the_United_States.svg https://en.wikipedia.org/wiki/Obesity_in_the_United_States#/...