This goes to the definition of causality. Ultimately you can never "prove" causality, even with the best designed experiment. You are allowed to make causal cla

This goes to the definition of causality. Ultimately you can never "prove" causality, even with the best designed experiment. You are allowed to make causal claims for non-RCT data. Lookup Bradford Hill criteria, just to start.