I always wonder how much publication bias exists in these fields: “Drug X isn’t that bad for you” seems like it might be a lot harder to publish than series of

I always wonder how much publication bias exists in these fields: “Drug X isn’t that bad for you” seems like it might be a lot harder to publish than series of papers ratcheting up the severity of some bad effects.