3 ms·

There is in fact a very strong correlation in both cases! ... just no causation in any direction in any case. =) * Any case out of the 2 cited by OP. (Thanks A

• Alex3917 16y ago
  If there were no causation then you'd expect to see the same J-curve for all causes of mortality, but that's not what you see. For example the more you drink the lower your risk of heart attack, regardless of how much you drink. There's actually very clear causation for certain causes of mortality and morbidity, it's just that the picture is a little bit murky when looking at total mortality and morbidity.