Wednesday, July 11, 2012

Simulation training vs natural history in LP battle royale

I read an article on the use of simulation in teaching lumbar puncture (LP) technique to residents by Dr. Simuni and her medicine colleagues at Northwestern University.  I thought it was a really interesting article and helps to add data to the idea that a curricular plan in medical education which includes deliberate practice and simulation does a really nice job of teaching learners a new skill.  It hints that this deliberate practice in a logical fashion is better than the traditional model.

I'm not so sure this paper really definitively answers the question that this practice is superior to the traditional training model.  In brief, the article pits final scores on a mastery checklist of first-year medicine residents who underwent a three-hour educational session including simulation to teach proper LP technique against neurology residents who were asked to simply do the LP simulation while being graded on their performance on the checklist.  I think this result may stand over time and additional studies, but I have a few problems with it.  First, the neurology residents were not shown the checklist.  I think this is a big deal.  I don't know the proper place to do that to get adequate controls, but essentially the medicine intern group was taught to the test.  It was deliberately pounded into their heads over the three hour session that these are the things they are going to be graded on.  That's what deliberate practice is all about.  It's about repeating something to get it right.  To my mind, that is teaching to the test.  The neurology residents weren't given time to familiarize themselves with the simulator (at least it didn't say they were).  They also weren't oriented to what they would be evaluated on, so of course they didn't perform as well on the checklist.  As they have likely done multiple LP's it might have also been easier to skip to inserting the needle in a simulated environment as it feels artificial, and it feels like this is the ultimate goal.  It might have been more useful to go to the bedside of the next LP these residents did and see if the 'real world' performance was different between the PGY-1 group and the neurology residents.  I doubt any neurology resident would forget to get informed consent in the 'real world' (but I may be wrong).  Maybe the simulation training in part, trains you how to take the final simulation exam.  I'm not saying that it was not a good idea to do the simulation training, I do.  I'm also not saying the checklist is invalid or has no practical applications, it does.  I'm saying that the PGY-1 group had the deck stacked in it's favor. 

I would also argue that the way I learned to do LP's was essentially with deliberate practice over time with multiple patient experiences.  When I was first starting, I had a senior resident or faculty over my shoulder giving me feedback on my technique.  Could this have been improved upon by adding a simulation session at the beginning of my training, absolutely!  But I don't know that this study really proves what they say it proves which is that traditional training is inferior to simulation, and neurology residents can easily be schooled by interns fresh from the simulation lab.  This is shown by comparing the neurology residents with the interns at baseline.  The neurology residents were all better just eyeballing the data.  I think if you had put the neurology residents through the training, they also would have achieved a higher level of mastery.  *That's a neurologist talking of course :)

I do want to say that I am a bit concerned at some of the mastery items the neurology residents missed (as were the authors).  The anatomy questions would have likely been taken care of by brushing up on the anatomy before the test, but you could argue especially a senior neurology resident should know that.  The authors were concerned about anesthesia, but that could have been a function of being in a simulator vs 'real world'.  It could also show how one of the schools has a local practice which is different from national norms.  The setting up of the tubes and manometer in a proper fashion is a bit vague to me, and I'm not sure I'd know what the proper position should be for that.  I wouldn't make that a make or break point on this procedure.  Not saying how I know this, but one can recover surprisingly well with the one handed method of unscrewing the caps in a pinch.

So this is a long response to the article as the editorial that accompanied it was trying to make the point that the traditional model was inferior and should potentially be reconsidered.  I don't think that is what this study showed.  I think it did prove that mastery level is attainable with a 3 hour simulation lab for PGY-1.  I'm not sure it really proves they are better than neurology residents.  They may be, but I don't think this was a fair assessment of that.