I guess if the benefits of a treatment are obvious you don’t need a lot in the way of research, or even science at all, to recommend it. But if the benefits ar

I guess if the benefits of a treatment are obvious you don’t need a lot in the way of research, or even science at all, to recommend it. But if the benefits are subtle or borderline, then it’s an interesting ethical dilemma. Which is the least unethical: withholding a potentially-beneficial treatment from a control group, or giving a potentially-non-beneficial treatment to everyone?