I guess there might also be confounders, as the patients aren't randomized. For example, if more senior consultants are males due to a gender bias in earlier ye

I guess there might also be confounders, as the patients aren't randomized. For example, if more senior consultants are males due to a gender bias in earlier years, they may be involved in more complicated cases, and so on. Some specialities are quite gender biased, which may also impact the outcome