However we don’t have evidence yet that grads are having to do fellowships en masse.
I'm fairly certain you wrote this off-handedly, but "grads having to do fellowships en masse" is beyond a bad job market. That's a catastrophe. That's a year of lost income, another year of accumulating student loan interest, another move away from friends/family, all without a real value-add outside of a few specific areas like brachytherapy or peds. I would argue that there is a line that constitutes a "bad job market" and it comes much earlier than near-mandatory fellowships.
Are we there yet? I would argue yes, but reasonable people can disagree. However, to say that we aren't headed in a very, very troubling direction with
increasing # docs per year and a
decreasing # number of fractions/indications is intellectually dishonest. This may turn around on phase III oligomet data, but if I were a med student evaluating my career options, I wouldn't predicate my future on that supposition.
Give med students an honest appraisal of the current state without supercharged rhetoric and let them weigh the balance of pros and cons. I think there are so many downsides to med onc, and I’d hate for students who otherwise would have loved the work in rad onc to instead choose heme/onc.
Agreed. I don't think any med student interested in rad onc should automatically do IM and heme onc instead, I just think it's important to encourage med students to have a good idea what the other options are out there.
To play devil's advocate, is it not possible that stats and quality of entering residents went down when the specialty collectively decided to increase spots annually, often by double digits?
Thankfully, we have data in the form of "Charting Outcomes in the Match" to evaluate this. Here it is:
2007
# Matched: 134 (US + independent)
Mean/Median Step 1: 235/236
25-7th %ile: 223-248
AOA: 24$%
PhD 21%
Mean abstracts/presentations/pubs: 6.3
2009
# Matched: 142
Mean/Median Step 1: 238/241
25-7th %ile: not reported
AOA: 35%
PhD 22%
Mean abstracts/presentations/pubs: 9.7
2011
# matched: 164
Mean/Median Step 1: 240/244
25-7th %ile: not reported
AOA: 31%
PhD 22%
Mean abstracts/presentations/pubs: 8.3
2014 --
this group was PGY4s during 2018 rad bio/phys testing
# matched: 177
Mean/Median Step 1: 241/248
25-7th %ile: not reported
AOA: 23.6%
PhD 23%
Mean abstracts/presentations/pubs: 12.2
2016 --
reported differently from prior years; NRMP required applicants to 'opt in' to report data; may skew data
# matched: 149 (lower than NRMP reported # of 185 matches)
Mean/Median Step 1: 247/251
25-7th %ile: not reported
AOA: 27.5%
PhD 24.8%
Mean abstracts/presentations/pubs: 12.7
2018 --
like 2016, this was self-reported data and sample may be skewed
# matched: 165 (lower than NRMP reported # of 188 matches)
Mean/Median Step 1: 247/253
25-7th %ile: not reported
AOA: 35.2%
PhD 20.8%
Mean abstracts/presentations/pubs: 15.6
Additionally, someone asked Dr. Kachnic at the ARRO seminar at ASTRO 2018 how the 2018 examinees did on the 'recycled' questions. Apparently, these recycled questions comprise something like 30-60% of the test every year. These re-used questions can provide perspective and year-to-year comparisons. Apparently the 2018 examinees did, on average, better on those reused questions than did the examinees who answered the same questions in 2017.
Taken together with the match data, I have trouble believing that as of 2016, our applicants are the driving factor behind worse exam performance.