Let the flame wars begin:
Fundamentally, we have a very noisy and limited, although substantial data set. It is unlikely that these data allow us to be confident in regards to the strength of any signal regarding the relationship between physician sex and a patient result that’s really vital that you the authors.
I am reluctant to stack methodological and statistical standards to produce support for such a change. There could be unwanted effects of accepting that arguments can be made for a good cause, even with such signs that it is poor.
Imagine if the authors had found the same modest differences in noisy data in the inverse way? Would they assert that we ought to maintain gender differences in doctor pay? Imagine if the writers concluded that lower pay which girls receive was correlated with reduced mortality and focus on another variant in all this sound? Would we then advocate that will decrease the pay of both male and female physicians in order to improve patient results?
See the rest at plos.