I disagree that that is the only way of constructing a study which is effective in investigating this question. I also don't accept that the fact an RCT is not viable would somehow imply that we should draw out conclusions from studies like the above.

Well if we can't run RCT's, then we need to rely on observational studies, right? Can you suggest a better way of investigating this question?

Nicotine patches or inhalers would be an option? Assuming it is the nicotine and not a different chemical.

Like again, without incredibly strong empirical evidence, no ethics board will recommend a study on nicotine.

That would be a pretty ignorant ethics board then, given that pure nicotine at normal doses has few side effects, and unless combined with a monoamine oxidase inhibitor (as it is in tobacco) isn't particularly addictive.