Observational studies

From Amit’s Thoughts

People who wear red shirts on the original Star Trek have a high mortality rate.

1. can we conclude that red shirts caused the high mortality rate?
2. can we solve the problem by having them yellow shirts instead?

Every time I see observational study results like "people who eat lots of beef get cancer" it reminds me of the Star Trek scenario. "People who do X end up Y" does not mean X causes Y!

There are lots of things in biology where A correlates to B, but it's not obvious whether A causes B, or B causes A. There may be confounding variables involved, where C causes A and C also causes B. Maybe it only affects some people. Maybe the effects are just coincidence. Maybe it's several things that have to happen together. I think people want to have a villain to blame, whether it's salt or sugar or fat or carbs or meat or sitting down or pesticides or seed oils or food processing or video games or nitrates or butter. But biology is a lot more complicated than that.

We want simple answers, and favor studies that match our existing beliefs. Be skeptical! In moderation, of course.

Labels: health