Except, does it bias the outcome at all? If you do a meta study of a thousand different medications that work, you’d discover the single most common ingredients are starch or cellulose. Clearly it’s worth taking either or both to stay healthy! Of course this ignores the fact that they’re just filler, the active ingredients are the tiny fraction of the total recipe that are actually pharmacologically active.
The presence of a trait doesn’t imply that it’s in any way an active ingredient, even a very little active. Post hoc analysis is bullshit.
They can be used to treat disease, but are often used to prevent disease. Statins in particular are used to prevent cardiovascular disease from developing.
As far as I know, high cholesterol only correlates with heart attacks and strokes. there are also people who have high cholesterol and will not experience a heart attack in their life.
The presence of a trait doesn’t imply that it’s in any way an active ingredient, even a very little active. Post hoc analysis is bullshit.