I am a medical epidemiologist and I am speaking from direct experience.
I did my MPH at Johns Hopkins before doing a medical degree, where roughly a third my classmates were MDs. Another third were nurses, the rest of the class being filled out with various other health professionals. I was the lone engineer. I found that as a graduate student, my MD classmates consistently overestimated their own statistical abilities, to the point where in group work/assignments, the joke was to not let the MD do the statistics.
In general, MDs from research oriented universities who have been publishing through med school, with the goal of a competitive match at a high prestige research departments have good working knowledge of basic biostatistics and research methodology. These are the exceptions.
Although it's been a while since I've done basic research and published out of US based institutions, my opinion remains the same. Most American MDs are practitioners, not academicians. Not even close. IME, most private practice MDs don't even understand the statistics they're reading in their own journals in their own fields.
And yes, the English system is what I was referring to. I favor the M.Med to D.Med/MD progression because better distinguishes academic vs practitioners credentials. I'll probably end up with a DrPH.
As for a one line statement on health outcomes being a bad example, we can engage on this subject, but it's my suggestion that if you don't read too much into the statement, you'll find yourself being less of a contrarian.