Actually, if you look at the figure at the end, amongst slow walkers, it makes no discernible difference whether they are normal weight, overweight, or obese (o

Actually, if you look at the figure at the end, amongst slow walkers, it makes no discernible difference whether they are normal weight, overweight, or obese (of course a difference might be evident if there were more data). So the known risk is: normal & slow = obese & slow > obese & brisk > normal & brisk The last two > comparisons are a bit uncertain individually, though the combined one (obese & slow > normal & brisk) is statistically significant.