Explaining the Anti-Compliance Confounding Factor

One of the biggest problems in trying to determine what is or is not a healthy lifestyle is the existence of confounding factors. One of biggest confounding factors is the tendency for people who lead healthy lifestyles in general to also follow lifestyles that are widely considered to be healthy. On the other hand, it's common for an observational study to show that X is healthy only to reverse course a decade later. For example, the Framingham study showed lower death rates among people who were close to anorexic. More recent data shows that the lowest death rates are among people that the Framingham study classified as mildly overweight. It's as though the supposedly-healthy users are actually following an unhealthy lifestyle in general.

We see a possible clue in a recent study from Austria which found that vegetarians tended to be in ill health:

Moreover, vegetarians are vaccinated less often than all other dietary habit groups (p=.005)

Hmmmm… Maybe the ‘healthy users’ are less vaccinated. If that is the case, we can expect headlines a decade from now showing ill health among people eschewing GMOs, trans-fatty acids, or high-fructose corn syrup.