He seems to be using small clinical differences over the population as a proxy to insinuate that the effects on the individual are small as well. They aren't.

He seems to be using small clinical differences over the population as a proxy to insinuate that the effects on the individual are small as well. They aren't. The average might be small, but the range is wide.