Thanks for sending the link to the paper on selecting residents. Many surgeons feel that choosing athletes who played a varsity sport—team or individual—in college is a good way to pick residents. With one notable exception, my limited experience is consistent with that idea. It's limited because there are not enough applicants (at least not to programs I ran) who are athletes. I have a post coming out soon about the subject of "grit" or conscientiousness and selecting residents who have high grit levels. A recent paper suggests that residents who drop out of surgical programs might have low grit levels.
The resident who wrote to me suggested trying to choose applicants who fit in. At first glance, the idea is appealing. However, the matching process can thwart that goal because the people you think will fit in may not rank the program highly. If everyone based their selections on who fit in best, there might not be women or minorities in many programs.
Teaching residents how to dictate operative notes is important for residents. The problem with allowing a resident to dictate a case is that the dictation is a legal document and cannot be removed from the chart, particularly if it is an electronic medical record. I have always felt that if a resident cannot coherently dictate a case, she probably did not learn how to do it and would not be able to do it by herself. Practicing off-line using speech recognition technology could overcome this problem. The resident could dictate a draft which then could be gone over with the attending thereby achieving the feedback which is a very important part of learning.
Regarding the best use of limited didactic time, I have no brilliant answers. In fact, I'm glad I'm no longer a program director and don't have to deal with this difficult question. One often overlooked factor in work hours limits discussions is that conference time has been quite negatively impacted since 2003.
Because about one-third of residents must go home early every morning means that there are no longer any afternoon conferences or rounds. Cramming 2 or 3 hours of didactic time into a single morning goes against many principles of learning especially if the sessions are boring lectures which do not engage the audience. Intermittent bursts of teaching and/or practice have been found to be better for learning than long single sessions. In addition, there is so much more to learn because of the expanding body of knowledge and mandates from the RRC and other regulatory entities.
I have written several posts advocating teaching residents how to think rather than memorize facts which are available on a smartphone. Here's one from 2012. However, this will require a top to bottom reorganization of not only the way residents are taught, but also the way they are tested.
Please comment if you disagree or have something to add.