As many commentors note, this is a comparison based on Sierra Leone MDs and assistants. Sierra Leone (as the article notes) is nearly at the bottom of the inter

As many commentors note, this is a comparison based on Sierra Leone MDs and assistants. Sierra Leone (as the article notes) is nearly at the bottom of the international "human development index", so very poor - there are not enough surgeons or even MDs to treat everyone. The surgery studied is an elective hernia surgery. In a wealthy country this could be done in a "hernia repair factory" and I recall, but won't find, a study showing that is the best case: the more practice the surgeon has, the better the outcomes. I would expect the same results here: if the MDs do not specialize in hernia repair, they will be beat by the specialist (even if not an MD). Very importantly, this was a RANDOMIZED clinical trial. That means that counfounding variables should be equally distributed by the randomization. You generally do not have randomization when you are comparing outcomes, e.g., between the Cleveland Clinic and the Mayo Clinic in the US. That is a great feature of the study. It gives me much more confidence that what they detect is a true difference, not caused by a factor like selective recruitment.