Haven't gone through the thread completely, but I think one of the PAs there posted a balanced viewpoint:
"The problem is you need more information than just the pass rate. I don't think that USMLE has ever published their exact scoring process, but I am guessing that its a dynamic scoring process similar to what is used on the PANCE where each question carries a certain weight. Depending on the question you can have a variable passing score. I am also guessing that the "experts" that developed this test off of what the USMLE provided did not have the experience to do this. This essentially makes the pass rate statistically questionable.
I would also like to see some demographics associated with this. There are a number of NP specialties represented here theoretically. It would be unusual for a Neonatal Nurse Practitioner whose patient population consists of those between 1 day and six months old to pass a test that is primarily based on adult inpatient medicine. Similarly given the number of pediatric questions on the example test, I would guess that someone doing adult outpatient medicine might have a hard time. Since they are testing NPs in areas that they are not trained in, essentially what they are measuring is the ability to pass the test. It calls into question the whole "certification" concept. "