How to Revive a Failed EHR Implementation

The more I talk with hospital executives, the more convinced I am that one way to improve the quality and acceptance of the EHR is being overlooked.

A hospital EHR is an incredibly complex system, a system that has the possibility of interacting with almost every person who is walked or wheeled into the facility.  Please permit me to make a somewhat outlandish statement in order to make a point.  If you spent more than $10 million for your EHR — by the way, that figure is less than what some hospitals spend on monthly maintenance for theirs — what do you expect of it?  Did you wind up installing nothing more than a $10 million or $100 million scanner or image management system?

What metrics enable you or the C-suite to conclude, “Job well done.”  Is it a 10 or 20% productivity loss?  Is ‘well done’ a breakeven on productivity?  At a minimum, by when should you achieve breakeven productivity — two years, three years?

Here are a few thoughts.  What if there was a way to take a very complex business problem — after all, EHR is basically going to retool the entire way business is conducted within the hospital — and make the problem simple?  T. S. Elliot wrote, “If I had more time, I would have written less.”  The beauty in problem resolution, the ah-ha moment, often comes from being able to spell it out in a sentence or to.  If I can’t articulate a problem on a single page, or draw a picture of it, it means I have not thought about it enough to understand it.  There is an entire branch of physics whose singular goal is to define the entire universe in terms of a few simple equations.  I think IT problems can be thought through in the same manner.  Why throw resources at an entire complex system when only a part of the system may need them?

What if your EHR has been in place for a year or more and your productivity is still below what it was when your chart room had more square footage than the archives of the Smithsonian?  If that is the case, you have probably done some or all of the following:

• Brought in the vendor to tweak the system
• Used internal resources to try to boost the speed
• Hired external system “tuners”
• Provided additional training

What did that effort deliver?  Probably very little.  Trying to get to Los Angeles from New York takes a long time if your flight path requires multiple stopovers, or if the route requires you to fly east, north, or south in order to move west.  If you watch people in the exam rooms use the system, you are likely to notice that the problem has more to do with mouse moves, clicks, loading, and navigation than it has to do with processor speed.  No amount of speed will speed up the parts of the system that require human intervention — find the spot on the screen for the next piece of information you want to read or enter, move the mouse, click enter, and repeat.

While watching the people in the exam room, do not forget to see what the patient is doing.  I always bring a book with me when I visit my doctors because I know there will be a lot of idle time while the doctor or nurse does not even know I am in the room.  I even told my cardiologist that perhaps instead of me driving all the way to Philly for my next exam, perhaps we could just conduct my next appointment on WebEx.

Here’s an idea whose only focus is on how to make the time spent in the exam room more productive, saving clinical time, increasing face-to-face time, and thereby making the EHR more productive.  Attach a very efficient EMR as a front-end or GUI to the EHR.  Large EHRs are designed to do many things.  The likelihood of a cumbersome EHR being nimble at the point when being nimble is of great value is slim.  People will balk at this idea and say it won’t work or that it can’t work.  I am willing to bet that the naysayers have never bothered to see how it might be made to work.