Bear with me, I'll make my point eventually.
I did a little retrospective "study" recently for want of a better word
, using my ePCR data for about a month (because clearly I have too much time on my hands where I'm currently working).
I put each patient into one of the following categories:
1) Reasonable to call an ambulance (based on what I believe a reasonable lay person would call an ambulance for).
2) Needed and ED, reasonable person should have got there by their own means.
3) Selfcare/GP.
Then for each pt I also said whether or not an intervention was performed. (I didn't include "just-in-case cannulas").
I excluded psych transports under police custody and IFT.
I know. Its hardly scientific. Nothing more than a though experiment. But the results surprised me none the less.
RESULTS: n=42, By my reckoning, 66% of jobs involved a pt for whom I would have called an ambulance. This surprised me. I thought it would be lower. Of those that could have gone by their own means, it was often not their fault that an ambulance had been called (home care nurses, nurse on call hot line, call to a specialist physician who refused to consult on the phone and told them to call and ambulance). The second thing that surprised me was that, of the remaining 34%, 74% fit into the "selfcare/GP" category. I thought there would be a lot more people from both "selfcare/GP" and "ambulance" categories that would fall into the "ED by their own means", but as it turns out there wasn't much between, "I legitimately need an ambulance" and "I have had a cough for 2 hours".
We intervened in 28% of pts. Most of whom fit into the "ambulance" category. I didn't count the advice we gave the selfcare/GP people, and the assessment and decision making that lead to them staying at home which I think is pretty important.
This made me feel better about that fact that I was feeling a bit like a taxi driver at the time (which was the purpose of having done it in the first place) and it also re-enforced a reasonably well established idea that quite a few of our pts needs primary care, not emergency care. What my little thought experiment showed me was that, rather than most of our patients being people we couldn't do anything for, many actually required our services as healthcare professionals in the emergency sense, and quite a few more might have benefited from the primary care advice we might have given them. I wondered if we could have done more, had we the expertise and scope and it of the paper I wrote in uni arguing for more primary care education.
The questions of "how much is enough" when it comes to education and scope, isn't just limited to the intensive care end of the spectrum. I think there are a lot of good arguments of the expansion of scope in a lot of areas. The rise of the noctor (to quote some idiot), has some legitimate concerns associated with it but I think there is a good argument for the expansion of scope in a lot of healthcare professions, to fill the gaps, to expedite and streamline healthcare and to improve outcomes in a lot or areas.
Back to Bernard and paramedics "killing" patients. Until such time (or if) some sub-group analysis is done on those patients who suffered cardiac arrest during the RSI trial is done, I don't know that the use of pejorative and emotive language like "killing patients" is useful. There are a raft of reasons why patients may have had a higher cardiac arrest rate in the pre-hospital arm, like survivor bias. We don't know what agents were used in the in-hospital arm of the study for induction: is it possible that hemodynamically unstable patients were inducted with ketamine rather than the fentanyl/midazolam that the paramedics were restricted to, thus sparing their perfusion further insult?
I think you've hit the nail on the head. Survivor bias.
To add to this is a further example of what I was saying about adding to scope and not subtracting. The addition of a paralytic probably improved prehospital intubation. Survivor bias aside, if the results of this study actually reflect some small "kill rate" on our part, I would hazard a guess that it might have something to do with a lack of induction options, not too many.