Posted by: Jack Santos
Those of you that know me or my background know of my healthcare IT pedigree – on both the insurance and the provider side. Keep that in mind when reading this post, because my comments can be easily misconstrued.
One of the key concerns for healthcare IT advocates (and detractors) is the automation of healthcare; kind of a care assembly line. This has certainly been a concern of some watchers of the debate going on at Capitol Hill.
This concern manifests itself with alerts and reminders in an Electronic Medical Record system, leading to complacency on the part of the clinician. I recently saw a similar phenomena happen when I took in my brand new 2009 car to the dealer for a perceived warranty issue.
Mechanics courses now focus heavily of use of computer based diagnostic testequipment. By connecting to the car’s on-board computer (OBC) system, a mechanic can measure and detect all sorts of things – alerts on component failures, out of spec operation, even speed and acceleration statistics (which would be a boon for auto insurers and rate determination – but that is for another post).
In my case, my cruise control on my new car was working erratically – sometimes over 30 MPH, sometimes under 40, sometimes never. After three trips to the shop, and numerous on-line lookups by mechanics for similar service recalls or problem reports (each time downloading codes from the OBC and diligently following alerts and repair protocols) they determined that the problem was an obscure part that rarely fails, especially in so new a car: an electronic brake light switch. It would take a month to get a new one from the factory.
What the mechanics failed to do in this entire diagnostic process was to actually LOOK at the patient (in this case, my car). Even a rudimentary look would have revealed that an after-market hitch harness for brake lights had been installed (by yours truly). The manufacturer of the hitch harness had incorrectly listed this make and model as compatible. The electronics in the harness (which was supposed to just extend the brake light signal to a trailer) interfered with multiple things in the car – including the cruise control. I removed the harness – and problem solved.
This automotive story is a perfect example of what I would call process automation complacency. Complacency that would be the result of relying totally on pretty (although busy) computer interfaces (like the one above), automated alerts, reminders, computer based knowledge bases – at the expense of rudimentary human-based diagnostics, or even common sense.
In this automotive case, the results were just annoying; in the case of back office processes, it could be an impact to the bottom line, or even financial reporting. But in healthcare - it could be a life lost.
Should that derail the Electronic Medical Record juggernaut? I think not. But it should make computer professionals, and healthcare executives, sensitive to all the other aspects of a healthcare delivery system – including training, and the cultural/sociological implications. Cars can be replaced – human lives cannot.