I would echo the above comment. I would also consider starting your own projects. This way you can do exactly what you are interested in and not have to jump onto somebody else's project that you may not care about. It'll also guarantee you a first author pub.
My focus is on using databases to conduct research that are already de-identified which cuts out the IRB and the worst part of the research process (data extraction). Several databases are free and are nationally representative of the U.S. population. BRFSS, NHIS, and NHANES are all examples. Another example of a database that is not free (maybe a few hundred dollars if you're a med student) but has very good data from Medicare is the national readmission database. You basically just need to look through the questionnaires from one of these databases, make a hypothesis, and see if it's been published before on Pubmed. For example, using BRFSS you could look at if smoking and physical activity levels differ among U.S. veterans compared to civilians in persons who are diagnosed with COPD. This is just one example of something I just searches on Pubmed. In order to get a project like this done, you'll need to reach out to a few statisticians at your institution (they're there, you just need to ask around) and ask if they'd be willing to perform the statistical analysis for you and in return they get to be a coauthor (you shouldn't be paying for statisticians fyi...).
On a side note, the DO/MS in med education will be a waste of your time unless you're actually interested in med ed. You won't get any significant publications out of it and would be better off spending the time it takes to do the extra classes writing up manuscripts and getting more pubs.