On reflection, I should have let this one through. Ordinarily, I'd just check the references myself and ask the submitter to provide their own in future.
On this occasion, I rejected it partly out of frustration at seeing multiple pending submissions without any references, and not wanting to encourage bad practices.
However, we should all be assuming good faith, so I don't plan on rejecting any more submissions purely on those grounds.

医学英和辞典, 中辞典, etc.
G n-grams: 831

I think it is perfectly reasonable to expect that someone submitting a proposed entry has 
done some checking beforehand, and includes details of that. It's a shared task, and I don't 
think it's appropriate to throw the whole verification process onto an editor.
I can usually check a term with a couple of copy/pastes, and in this case the term checks 
out, but I certainly prefer that it arrives with some sort of case for inclusion.

https://www.edrdg.org/jmdictdb/cgi-bin/entr.py?svc=jmdict&sid=&q=2843916.1

Rejected once previously because "no references". 
But even if there had been a reference or two, wouldn't you have double-checked them? 
So why couldn't you just check your usual references or at least googled it the once? check it against the n-grams?
Doing that and rejecting it based on the word not being common enough or whatever, that I'd get. But rejecting a WORD, as a dictionary editor, because you're not happy with the SUBMISSION? Doesn't make any sense to me.