Probably the FDA, however IRBs can difficult to deal with as well. Here are some examples from my experience, albeit they're mostly non-emergency cases. Regarding the FDA, a study I was working on with a PI, looking at sternal wound healing, was pretty much all ready to go (to the point that we were already identifying patients to consent from the next week's surgery schedule).
However, out of the blue, the FDA suddenly decided they wanted to re-review our use of a device used to extract certain components of blood and bone marrow that would be applied to the wound during closing. This device is already FDA approved for very similar/near identical purposes in ortho procedures. However, because we were using the products extracted from the device on bones where it really hasn't been used that often, that it would have to be deemed an "experimental device". Which, in the PI's and my opinion, is bs as we had already been given the go-ahead by them almost a year prior....and we're still waiting (after nearly a year) beings the FDA, like all government agencies, loves to take their time. On a personal note, this really sucks, because I would've received a stipend for this study on top of my salary, gotten to run the data, and received authorship; but beings I'm done in three months that won't be happening
🙁.
Regarding the IRB, I'm currently working on a study looking at longitudinal health outcomes from certain age populations of patients with certain CV diseases, which almost got dumped because of our IRB. Their reason? They didn't approve of the way we were going to identify potential participants (i.e.,
running reports from our EMR of patients admitted with these diagnoses). They (actually it was just one member) felt it violated HIPPA because we would be entering patients' charts without their consent to see if they qualify. However, this is how we identify patients for nearly ALL of our studies unless a physician refers a patient. It took three meetings with the PI, my boss, the IRB, our research attorney, and myself to get them (well, him) to budge. I still have no idea why he decided to bring up this issue on this one trial when this is what we do all the time.
I'm less experienced with these issues in emergency type trials, as all of the ones I've worked on were already up and running when I started my job. However, there have been three trials since I started that
were all given the boot by our IRB even though they had FDA approval, because they felt those trials ran the risk of causing undue duress on the patients and their families under very stressful situations (and to be honest I sided with the IRB in these cases). I think the issues with IRBs largely depends on the type of hospital (i.e., community v. academic), the general atmosphere/way day-to-day clinical activities generally run, and whether or not you have to gain approval from your local IRB or can use a centralized IRB.
Sorry to inundate you with clauses and run-on sentences (just the way I tend to write), but I hope I was able to adequately touch on your question. As for my job, I enjoyed doing research in undergrad, and wanted to get out of working as a CNA after I graduated while staying in the medical field. I saw the opening and applied. The standards for getting a job in clinical research are unfortunately, surprisingly low. Case and point, one of my "most highly regarded" co-workers argued with me for damn near an hour that a carotid endarterectomy is the surgical removal of your carotid arteries

- if only they would do that to her. Also, the run was good
🙂.