Getting something published is noteworthy and important in itself regardless of the topic and where the research is being done.
I think it's easier to churn out chart review papers.
I can't say I disagree with this statement, but I will clarify with this:
As someone who reviews apps for residents and fellows, I would rather see one solid research paper with original durable data that a student spearheaded opposed to 5 case reports or even a review.
Quality over quantity in general.
The impact of the project will determine if it is publishable or not. I am the current faculty mentor on the QI project for our fellows and I think it will be publishable. It is something that other places would want to know about, but I would guess it wouldn't be much more than a 'letter' type publication. If we had more than a year to work on it, I think we could surely flesh it out to a solid manuscript though.
If you want a full blown project/manuscript, you'll likely need to latch on to a mentor who is connected.
A lot of QI projects involve surveys. There is a lot to survey theory (it's not just throwing questions on survey monkey). Not all need surveys, but it a good way to show before and after changes (through workplace satisfaction, process improvement, etc).
The best QI projects improve a real life problem that providers (or others) have often. So it may help to talk to a few providers in your field of interest to learn about some things that can be improved upon. An example would be in surgery, patients calling back after surgery with wound infection concerns. The intervention would be creating a brochure with pictures of healing wounds and sending it home with pts on d/c. You would take metrics on number of calls about wounds before and after the intervention (that one wouldn't really involve a survey probably). That is an example of something that you could probably publish if you could get enough data. It improves quality (giving pts better health information), cost (less time spent by providers addressing those concerns and being more productive), and even access a little bit (not having to call for a concern in the first place) (the three main big picture metrics of healthcare).
I don't think it matters if it's community or academic.
I would still hold a solid bench research paper of similar quality above a QI project personally, but I would certainly be impressed with a 1st/2nd year med student who spearheaded a good QI project that was published.
Disclaimer: The following is totally my personal opinion
Bench > Clinical (clinical trial > cohort > case control) > QI > Meta analysis > review (chart > lit) > observation study > case series with review > case report
A start to finish solid translational project to me is the ultimate, though.