Howdy. I'm a FM-trained flight doc in the USAF at a base on the new DoD EMR MHS Genesis (which appears to be a budget version of Cerner). The transition has been rough. AHLTA sucked hardcore 10 years ago, but it was almost a functional EMR by the time we switched.
We were promised the world. Well, maybe not the world, but we were promised more than we got. One of the things we were told was that allergies and immunizations would absolutely 100% be imported. The funny thing is, Cerner managed to import the immunization records, but it can't utilize significant chunks of them, so it still recommends a lot of immunizations that are not indicated (in particular, it has no idea what "titers" are and why they would matter).
Our marching orders are to manually enter all positive titers as "immunization not given", which is a click-intensive workaround to a problem that we were promised wouldn't exist. As I was entering yet another positive varicella titer today (thanks pox parties), I was reminded of an XKCD comic about digital data and the myth that it never degrades.
I was struck by the realization that, in 5-10 years, we will likely switch EMRs again, and all the hard work I'm putting in to get Genesis manually synched up with AHLTA (our old EMR), CPRS (the old VA system), HAIMS (the server for scanned documents), CHCS (most of which can be found in AHLTA, but not necessarily everything), Essentris (our old inpatient system), and ASIMS (our immunization and readiness system) will be flushed.
I've always drank the cool-aid on granular digital data. When we were using AHLTA, I clicked the individual ROS and PE boxes (and yes, sometimes I even left the TSWF to do it old school in the even-clunkier SO module) because the line we always heard was "something something data extraction yada yada clinical research blah blah population health". I have now watched years of my diligent work get flushed because no one ever built a tool to actually look at or do anything with that data.
Everyone's bad-practice workarounds contributed to EMR clutter/bloat that is one of the few things that can still be easily seen when viewing old records (if I had a nickel for every patient who carries the chronic diagnosis of the "patient education: diet" disease, I'd have a lot of nickels).
So where do we go from here? I trusted the system and made my data granular and Cerner took a big old dump on that work. How do we increase the odds that our work in the current system will be useful and usable in 10 years?
Source: Original link