We know that training students in clinical procedures is a difficult issue. Even if we ignore the problem of too many students competing for too few procedures, concerns about liability, supervision, reimbursement and technical problems remain. Virtual reality (VR) simulators have been touted as the next generation in training devices, but a literature scan reveals minimal efforts to truly validate the efficacy of this training. Last week, we discussed a study that attempted to validate the use of VR training in a clinical setting and discovered that performance did improve after training with the simulator, but there were limits to the learning. The majority of the increase in skills occurred during the first three simulated attempts. Continued practice resulted in only minimal performance benefits. When looking at the improvements in performance between the pre-test and post-test, the gain was relatively small. The investigators concluded that although the VR trainer did work, the ability to transfer the skill acquired through training was limited.
That means EMS providers need access to real patients to achieve excellence in procedural skills. The question: How many patients do we need to see and how many times must a procedure be performed to ensure competency in that technique?
For the sake of this discussion, I'd like to focus on endotracheal intubation. This is a relative low-volume, but incredibly high-risk, procedure that lies at the core of our lifesaving efforts.
The literature provides some interesting perspectives. A study from Australia notes that the odds of success of orotracheal intubation in a training setting are intimately linked to the type of training device used. In this work, the chance of a successful intubation fell by about 50% when a student was moving from one trainer to another. Interestingly, the same study found that a student learns more from one successful intubation than from 12 failures. I can't help but think that because every patient is different, every patient is their own training device. For each patient, the individual odds of successful intubation fall. And I'd hate to think that one has to miss an intubation 12 times to learn to get it right.
(Reflecting on the noted venipuncture study, VR training has been proposed as a valuable adjunct to learning intubation as well. When applied to patients, however, the above data reinforces the impression that its utility may be limited.)
If we need humans to train humans, the next question is how many live patient procedures are required to demonstrate adequate skill and knowledge. A Canadian work tried to assess the number of successful intubation cases required to assess competence in non-anesthesiologists. Twenty "intubation novices" performed a total of 438 procedures. Performance was compared to a standard established by anesthesiologists. Statistical modeling showed that 46 attempts were required to ensure a 90% chance of a "good" intubation.
If we extrapolate these results to our current training schemes, the results trigger a few alarms. I would venture that no training programs or EMS agencies would require a paramedic to perform nearly 50 intubations before considering them "competent" in the procedure. The latest revision of the Department of Transportation EMT-P Curriculum suggests that the student should be able to successfully intubate at least five live patients during the training program. In reviewing the post-graduate "clearance" requirements for EMS agencies, the rare service requires three or more intubations as a demonstration of competency prior to independent practice. The discontinuity between research and practice is obvious. If a medic gets 10 tubes under his or her belt during education and field orientation, it's still only 20% of the intubations needed to ensure a 90% chance of success (and I'm not sure any of us consider a 10% risk of intubation failure acceptable).
So as people with a vested interest in the outcome of the debate, we are going to question these results. We could quibble about what is considered a "good" intubation. We could say that an anesthesiology standard is far too high for our trainees and that since we're doing the procedure under emergency conditions the only thing that matters is success or failure. Perhaps a good intubation is simply one that gets the tube in the right place, regardless of elegance or grace, kind of like the only aircraft landing that matters is the one you walk away from.
I can understand the objections, but I'm not sure I buy them. With the variety of airway techniques and adjuncts available to prehospital personnel, I think there are very few excuses for a "good" intubation rate less than 90%. (I'm making the assumption that the patients selected for intubation had appropriate cause and no gross contraindications.) And I just can't buy that physicians in training require 50 intubations to get it right but paramedics figure it out in less than 10. Granted, there may be some differential brainpower going on here (and I'm not even going to venture which side I favor), but the odds just don't work out.
So how much training is enough? The truth of the matter is that one can never learn enough. No matter how long a student spends in the classroom, no matter how intense the degree of practice, the very nature of clinical care means that you can't know everything. Eventually you have to turn someone loose. But because some minimum demonstration of competency is required, the battle rages on.
Suffice it to say that I don't have a magic answer. I think the ideal number of procedures required prior to independent practice of prehospital care is different for each individual. It is also not limited to the technical performance of a procedure, but also intimately linked to the overall knowledge base of the paramedic as well as less tangible factors such as maturity, adaptabilty and situational awareness.Clearly, however, something is wrong when we sacrifice the findings of educational research for expediency and economy. If we want to do intubations well, there can be few shortcuts. Let's resolve to acquire both the knowledge and the numbers to get it right.