I don't recall off the top of my head, and I think it was only NPs and not PAs that took the test.
A little looking:
http://www.ama-assn.org/amednews/2009/06/08/prl10608.htm
Looks like the one commonly mentioned had 45 students from what is considered one of the better DNP schools. So no, it isn't a huge N value. Although someone with a little stats savvy could probably easily tell us that @ n=45 and n=20k for residents passing >95%, the numbers are probably statistically quite different and increasing the n value wont change anything. A low n is a common issue with experiments, but that doesn't mean that outcomes can always be thrown out until everyone is satisfied by the population size. As long s the math works out conclusions can be made
I also reject your arguments that they didnt have enough time to prepare - most residents use their clinical training as the time to prepare and review for a couple days before taking the test. It's somewhat naive to just assume all these nurses came in without having looked at anything and were probably also hung over and hungry or... whatever
🙄. I'd wager they studied a little bit longer than the average resident as they knew this was the opportunity to show they are equivalent to physicians. The test was also watered down from the actual usmle3 which doesn't help.
Personally, I would like to see a study where a program is blinded to the status of its "residents" and gets a mixed bag PGY1s and NPs and then we compare the PD evaluations. The logistics and ethics of this may make it impossible, but IMO there are not good ways to compare the two other than trust the correlation between training and expertise. As it stands now, the majority of DNPs work under physicians anyways in some capacity and sh** often rolls uphill such that poor outcomes due to NP incompetence will be hidden by work done by the entire medical team.
The other way to do it is for physicians to completely wash their hands of patients who get DNP care and that way when something gets missed or screwed up it gets recorded rather than cleaned up by a competent provider. The current papers put out by nurses look at outcomes for all patients in outpatient settings which really isnt a valid way to look at it. They want to run a cohort of DNP vs MD patients when what we need is a case controlled study, or a kind of a hybrid. Patients with disease X are more/less likely to receive proper care from _____. That study has yet to be done.