Yeah, I really wish we had solid data on this. If I ask 7 different people about this, I will get 7 different answers.
Been dedicating my time to manuscripts, but for every manuscript I publish, my peers publish 2-4 case reports. Or for every poster I get, they get a case report. I'm starting to think of that old quote "PDs can't read, but they sure can count". Some people say yes, some say no, some say it depends on the specialty/location.
It 100% depends on specialty/location, because it is a completely different ballgame and different emphasis and also reflects (for better or worse) your school's prestige. A random mid-tier ortho program may indeed be just fine with a high quantity of publications to prove that you're "committed" to the specialty. But if you want to go to an ivory tower, and you're not already in the ivory tower club, they're going to know why it's important for you to go there and what you're going to bring the program. Those PDs can definitely read. If you're just bringing a bunch of case reports, that isn't anything--they want people who come out of their programs to go on and "be somebody." Sounds snooty, but it is what it is.
I think there actually is some solid data if you want to look for it. Just for one example, let's look at the intern class at MGH:
Who We Are
They report that a full 40% have a PhD or Masters--meaning they explicitly are recruiting applicants with a formalized background in research or some other quantifiable skill. Of the ones who don't have another advanced degree (and were not prelim only), here is a sampling of their backgrounds:
Harvard x4
Brown
Boston U
Yale x2
WashU x2
Vanderbilt
Hopkins
Stanford
Duke
Cornell
Penn
Penn and did a research year
Emory who spent 2 years doing research at the NIH
UNC and played minor league baseball
Worked for NASA before med school
Had a career in book publishing before med school
Other x10
Clearly, somebody from "other" is getting in (though I'm being fairly arbitrary in what counts as "other"--for example, one of those is UVA, which is obviously still a very good school). And this is just based on my skimming the bios, I'm sure those 10 likely have something in their background that helped them stand out--and a number of them, FWIW, did have a compelling background story that isn't easy to quantify. But my underlying point is that at these top programs, they really value substantive research, and if you don't have that then they tend to gravitate towards the applicants from a shiny medical school. I'm sure you could do a similar exercise at other brand name programs.
Which again leads to the question--why is it so important to go to a "top tier" IM program?