I'm quite confident that people admitted to institutions for suicidal thoughts are much more prone to suicide. But it doesn't actually matter to my point. It's

I'm quite confident that people admitted to institutions for suicidal thoughts are much more prone to suicide. But it doesn't actually matter to my point. It's a very reasonable hypothesis, true or not. It needs to be addressed as a potential confounding factor, either by showing it's not true, or by showing that the size of the observed effect is greater than the effect from the confounding factor.