Research during elective rotation

[Apoplexy__]

Hey, I'm an MSII with a question: How logistically feasible is it to conduct a cohort study during an elective rotation? Specifically, I have a couple cool ideas for a retrospective study examining risk factors for infective endocarditis. Is this something a medical student can just decide to do during a 4-week time period, or am I just being naive? I was hoping that most of the work of this type of study would be collecting data from a wide variety of health care institutions, which I could do slowly over time during other rotations, then actually do the data analysis and composition of the paper during the 4 weeks I'll get.

Realize I have next to no experience in clinical research and am completely in the dark about the logistical hurdles I'll potentially face. Any help/advice is appreciated.

---

Members don't see this ad.

 

[Apoplexy__]

Bump, I'd still love any input. Let me know if I'm in the wrong forum.

 

[Petypet]

It is not feasible if you have no experience with the process. A lot of the administrative stuff can be overwhelming your first time through, such as getting IRB approval (can take weeks or months), learning how to do chart reviews (where to actually find the data), then learning how to report the data.

My first time through clinical research (having a science PhD) took about 15 weeks from IRB approval, to doing the manual labor of pulling charts, to compiling the results, having a biostats guy look at the results, then to reporting the results (abstract + compiling a journal article). You also need to have a PI (MD or PhD) that can mentor you with the project, since you can't really do anything from the research to publishing without one.

My recommendation would be to hatch on with a PI in the field of your interest and get involved with some of their going projects (retrospective or even a prospective) to see how the process works. Once you are through one project with a PI, pitch your idea to them and 1. see if they will back it and 2. if its possible in the time frame given your rotations.

 

[Apoplexy__]

It is not feasible if you have no experience with the process. A lot of the administrative stuff can be overwhelming your first time through, such as getting IRB approval (can take weeks or months), learning how to do chart reviews (where to actually find the data), then learning how to report the data.

My first time through clinical research (having a science PhD) took about 15 weeks from IRB approval, to doing the manual labor of pulling charts, to compiling the results, having a biostats guy look at the results, then to reporting the results (abstract + compiling a journal article). You also need to have a PI (MD or PhD) that can mentor you with the project, since you can't really do anything from the research to publishing without one.

My recommendation would be to hatch on with a PI in the field of your interest and get involved with some of their going projects (retrospective or even a prospective) to see how the process works. Once you are through one project with a PI, pitch your idea to them and 1. see if they will back it and 2. if its possible in the time frame given your rotations.

Thank you, sir. Very helpful.

 

[mdeast]

Thank you, sir. Very helpful.

Totally agree with this post. It will really depend on what sort of retrospective cohort study your doing (is this from a national, de-identified database? is this from your hospital's EMR?).

In terms of the IRB process, oftentimes you can find a faculty member who is already conducting similar research for which your project may fall under an existing IRB (especially if they already have a defined cohort, either retrospective or prospective, that you could use). This could be potentially very helpful as they may have already done a lot of chart review and compiled data into a de-identified database that may be relevant to the questions your asking.

Secondary analysis of de-identified databases don't require IRB approval (you can usually apply for a waiver that gets approved much quicker...i.e. matter of days).

 

[Apoplexy__]

Totally agree with this post. It will really depend on what sort of retrospective cohort study your doing (is this from a national, de-identified database? is this from your hospital's EMR?).

In terms of the IRB process, oftentimes you can find a faculty member who is already conducting similar research for which your project may fall under an existing IRB (especially if they already have a defined cohort, either retrospective or prospective, that you could use). This could be potentially very helpful as they may have already done a lot of chart review and compiled data into a de-identified database that may be relevant to the questions your asking.

Secondary analysis of de-identified databases don't require IRB approval (you can usually apply for a waiver that gets approved much quicker...i.e. matter of days).

Great input, thanks. My idea is to look at risk-reducing factors for infective endocarditis, and I'm pretty sure I'm going to need a lot of case data to sift through to find qualifying cases. I'm just an MS2 so I have no hospital affiliation, but a national de-identified database sounds like exactly what I need. How does a medical student go about getting access to that?

I'll check around for professors doing clinical research. I've managed to snag a meeting with a faculty member later this week though, and will pitch some ideas to him. I'm not sure how many resources he or anyone other faculty will have, though...