I continued to find myself dulled by the confidence of many executives regarding how EHRs and ACOs (Accountable Care Organizations) are the way of the future.
Not that they are not, the way of the future, that is. My quibble has more to do with how one defines “future”, as though it may be the day after tomorrow, or the year after next. I also am troubled by the notion that implementing EHR throughout a hospital, and subsequently layering ACO processes and systems over that will be quite difficult.
So, how does one define difficult? My benchmark of a difficult process is folding a fitted bed sheet and trying to get it to look like it wasn’t folded by a third-grader. With that as a benchmark, as we think about the large healthcare provider business model I think it is very important to get our arms around just how difficult the task of implementing EHR and ACO are, and how far into the future it will be before things are functioning as they should.
It is never as simple as it looks. Let us remember the number of EHR failures for hospitals who think they finished implementing EHR is somewhere north of fifty percent. Just because the EHR vendor has left the building does not mean the EHR implementation is over. Many hospitals face years of EHR clean-up (disaster recovery).
For most hospitals, implementing EHR is the proverbial round peg in a square hole. No matter how you do it, neither the peg nor the hole will ever be the same.
Then you look at fitting an ACO model to that.
The only way ACOs will work, and work is what you want them to do, is to first stabilize the disrupted business processes of your EHR implementation.
Let’s assume that before you implemented EHR your hospital’s processes were somewhat less than one hundred percent effective—let’s call it 80%. Because the hospital effectiveness measure is below one hundred percent, and because nobody gets EHR fully correct, we should make a similar assumption about the efficacy of a large EHR implementation. Let’s say that when all is said and done EHR comes in at 80% of possible effectiveness.
To that we are going to try to add full ACO capability, whatever full ACO capability means. Now, because the hospital wasn’t fully effective, and because the EHR wasn’t fully effective, it stands to reason that the implementation of ACO will be ineffective, say 80%. If you allow me to take an overly simplistic shortcut to measure the upper limit of the ACO’s effectiveness we might peg the upper limit at somewhere near fifty percent. (80% * 80% * 80% = 51%)
While overly simplistic, the analogy points to an interesting result. Perhaps fifty percent is reasonable if you have a “glass half full” personality. If you are the “glass is half empty” type, one will conclude that implementing EHR and ACO on top of hundreds of ineffective business processes has made the organization considerably less effective than it was before the organization began to “fix it.”
What is one to do? Plan. While that is overstating the obvious, it is not meant as a trite answer. It is the right answer provided one frames the right question. The right question—plan what?
Therein lays the complication. Most are inclined to frame the question as follows, “How do we overlay an ACO on to our existing (or soon to be existing) EHR model?” Wrong question. A question with a higher probability of yielding a successful answer is “What do we need from an ACO?” This question then allows you to change the organization and develop systems to support the organization.
Answering the previous question requires you to strip functionality away from the organization to make ACO fit. This approach may make it fit, but it will not make it work.