The ambiguity you rightfully and correctly point out will not vanish. The reason is not so much conflicting evidence, but the seemingly unsinkable superstition that we can translate the results of group-based research to an individual patient. We can’t. The RCT, no matter how useful it is for clinical research, condenses its results into an “average patient”. The problem is that this average typically does not represent any of individual trial participants. Our major error lies in assuming that this mismatch miraculously disappears in the clinic. That’s why, in my post on aspirin, I explained how to weigh one’s individual risk for CVD events against one’s individual risk for Aspirin related bleeding, and make an individual decision based on those results (https://medium.com/read-or-die-hq/about-aspirin-protects-your-heart-or-hurts-your-health-2b757c4f3b30).