I would like to see a study like that controlled for cardiovascular exercise.
I am also concerned about the validity of the theory because of the trend stall in figure 2 at ages 80-89 and the stark reversal in the 90-93 age group, where high-high was at almost low-low levels.
It looks to me like there is a good chance of some unaccounted confounding variable, even beyond the possible effects of exercise (which would not explain the 90-93 group).
I am also concerned about the validity of the theory because of the trend stall in figure 2 at ages 80-89 and the stark reversal in the 90-93 age group, where high-high was at almost low-low levels.
It looks to me like there is a good chance of some unaccounted confounding variable, even beyond the possible effects of exercise (which would not explain the 90-93 group).