[asdkjh345fd]

>can't we simply compare the populations that consume a lot of omega-3 ( fish primarily ) vs those who don't?

That's how we came up with the hypothesis in the first place. But that's all such a study can do, give us a hypothesis. There are too many confounding factors (smoking rates, alcohol consumption, poverty, environmental toxins, ethnicity, other dietary differences, etc) to answer the question of omega 3 relevance. We need studies like this one, only with objective measurements instead of asking subjective questions to draw conclusions.