> Maybe people who have a more sedentary lifestyle also tend to come from a lower social background My favorite confound is when Pauling convinced many reasona

> Maybe people who have a more sedentary lifestyle also tend to come from a lower social background My favorite confound is when Pauling convinced many reasonably educated higher SES people that vitamins were a miracle cure-all. Suddenly multivitamins did correlate with higher health outcomes! (Until later studies untangled this and found little broad effect, aside from specific vitamins during pregnancy or specific deficiencies.) Sort of a pop science self fulfilling prophecy. Or when we do things we think are smart, sometimes we pollute the evidence.